Citation
Regulation of soybean nodule gas permeability and nitrogen fixation rate

Material Information

Title:
Regulation of soybean nodule gas permeability and nitrogen fixation rate
Creator:
Weisz, Paul Randall, 1952-
Publication Date:
Language:
English
Physical Description:
vi, 113 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Agronomy thesis Ph. D
Dissertations, Academic -- Agronomy -- UF
Nitrogen -- Fixation ( fast )
Soybean ( fast )
Nodules ( jstor )
Oxygen ( jstor )
Alkynes ( jstor )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1986.
Bibliography:
Includes bibliographical references (leaves 108-112).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Paul Randall Weisz.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
029846052 ( ALEPH )
15641497 ( OCLC )

Downloads

This item has the following downloads:

regulationofsoyb00weis.pdf

EX80M1I93_7JWW26.xml

regulationofsoyb00weis_0069.txt

regulationofsoyb00weis_0006.txt

regulationofsoyb00weis_0107.txt

regulationofsoyb00weis_0019.txt

regulationofsoyb00weis_0096.txt

regulationofsoyb00weis_0039.txt

regulationofsoyb00weis_0112.txt

regulationofsoyb00weis_0114.txt

regulationofsoyb00weis_0029.txt

regulationofsoyb00weis_0065.txt

regulationofsoyb00weis_0106.txt

regulationofsoyb00weis_0058.txt

regulationofsoyb00weis_0016.txt

regulationofsoyb00weis_0043.txt

regulationofsoyb00weis_0007.txt

regulationofsoyb00weis_0115.txt

regulationofsoyb00weis_0118.txt

regulationofsoyb00weis_0084.txt

regulationofsoyb00weis_0083.txt

regulationofsoyb00weis_0002.txt

regulationofsoyb00weis_0048.txt

regulationofsoyb00weis_0034.txt

regulationofsoyb00weis_0117.txt

regulationofsoyb00weis_0097.txt

regulationofsoyb00weis_0053.txt

regulationofsoyb00weis_0070.txt

regulationofsoyb00weis_0113.txt

regulationofsoyb00weis_0078.txt

regulationofsoyb00weis_0081.txt

regulationofsoyb00weis_0102.txt

regulationofsoyb00weis_0035.txt

regulationofsoyb00weis_0071.txt

regulationofsoyb00weis_0111.txt

regulationofsoyb00weis_0031.txt

regulationofsoyb00weis_0074.txt

regulationofsoyb00weis_0050.txt

regulationofsoyb00weis_0108.txt

regulationofsoyb00weis_0047.txt

regulationofsoyb00weis_0025.txt

regulationofsoyb00weis_0041.txt

EX80M1I93_7JWW26_xml.txt

regulationofsoyb00weis_0054.txt

regulationofsoyb00weis_0027.txt

regulationofsoyb00weis_0095.txt

regulationofsoyb00weis_0086.txt

regulationofsoyb00weis_0020.txt

regulationofsoyb00weis_0021.txt

regulationofsoyb00weis_0082.txt

regulationofsoyb00weis_0088.txt

regulationofsoyb00weis_0022.txt

regulationofsoyb00weis_0091.txt

regulationofsoyb00weis_0036.txt

regulationofsoyb00weis_0062.txt

regulationofsoyb00weis_0085.txt

regulationofsoyb00weis_0087.txt

regulationofsoyb00weis_0066.txt

regulationofsoyb00weis_0004.txt

regulationofsoyb00weis_0075.txt

regulationofsoyb00weis_0049.txt

regulationofsoyb00weis_0073.txt

regulationofsoyb00weis_0045.txt

regulationofsoyb00weis_0056.txt

regulationofsoyb00weis_0079.txt

regulationofsoyb00weis_0101.txt

regulationofsoyb00weis_0064.txt

regulationofsoyb00weis_0100.txt

regulationofsoyb00weis_0046.txt

regulationofsoyb00weis_0011.txt

regulationofsoyb00weis_0098.txt

regulationofsoyb00weis_0089.txt

regulationofsoyb00weis_0015.txt

regulationofsoyb00weis_0055.txt

regulationofsoyb00weis_0092.txt

regulationofsoyb00weis_0024.txt

regulationofsoyb00weis_0063.txt

regulationofsoyb00weis_0008.txt

regulationofsoyb00weis_pdf.txt

regulationofsoyb00weis_0059.txt

regulationofsoyb00weis_0038.txt

regulationofsoyb00weis_0110.txt

regulationofsoyb00weis_0042.txt

regulationofsoyb00weis_0080.txt

regulationofsoyb00weis_0099.txt

regulationofsoyb00weis_0028.txt

regulationofsoyb00weis_0103.txt

regulationofsoyb00weis_0061.txt

regulationofsoyb00weis_0105.txt

regulationofsoyb00weis_0060.txt

regulationofsoyb00weis_0017.txt

regulationofsoyb00weis_0119.txt

regulationofsoyb00weis_0023.txt

regulationofsoyb00weis_0109.txt

regulationofsoyb00weis_0076.txt

regulationofsoyb00weis_0057.txt

regulationofsoyb00weis_0104.txt

regulationofsoyb00weis_0068.txt

regulationofsoyb00weis_0120.txt

regulationofsoyb00weis_0051.txt

regulationofsoyb00weis_0067.txt

regulationofsoyb00weis_0032.txt

regulationofsoyb00weis_0009.txt

regulationofsoyb00weis_0001.txt

regulationofsoyb00weis_0030.txt

regulationofsoyb00weis_0040.txt

regulationofsoyb00weis_0037.txt

regulationofsoyb00weis_0010.txt

regulationofsoyb00weis_0003.txt

regulationofsoyb00weis_0033.txt

regulationofsoyb00weis_0090.txt

regulationofsoyb00weis_0072.txt

regulationofsoyb00weis_0005.txt

regulationofsoyb00weis_0093.txt

regulationofsoyb00weis_0116.txt

regulationofsoyb00weis_0077.txt

regulationofsoyb00weis_0052.txt

regulationofsoyb00weis_0094.txt

regulationofsoyb00weis_0000.txt

regulationofsoyb00weis_0026.txt

regulationofsoyb00weis_0012.txt

regulationofsoyb00weis_0018.txt

regulationofsoyb00weis_0014.txt

regulationofsoyb00weis_0044.txt

regulationofsoyb00weis_0013.txt


Full Text













REGULATION OF SOYBEAN NODULE GAS PERMEABILITY
AND NITROGEN FIXATION RATE















By

PAUL RANDALL WEISZ





















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY




UNIVERSITY OF FLORIDA




1986















ACKNOWLEDGMENTS


I would like to thank Dr. Thomas Sinclair for allowing me to work on this project and to undertake this research. His encouragement and tenacious optimism have been a foundation of support without which this work would not have been accomplished. I would also like to thank Mr. George Drake for his technical advice and dedicated assistance with the instrumentation and computerized equipment.

I would also like to thank my wife for keeping me sane in those bleak moments during this research when the data appeared hopelessly unintelligible.

































ii


















TABLE OF CONTENTS



PAGE


ACKNOWLEDGMENTS.......................................................... ii

ABSTRACT........ .................................................. v

CHAPTERS

I INTRODUCTION........................................ 1

II LONG-TERM EFFECTS OF ALTERED OXYGEN CONCENTRATION
ON NITROGEN FIXATION OF INTACT FIELD-GROWN SOYBEAN........... 7

Materials and Methods..................................... 9
Results......................................................... 11
Discussion............................................... 14

III NON-STEADY-STATE NODULE RESPIRATION AND NITROGEN FIXATION IN RESPONSE TO ALTERED OXYGEN CONCENTRATION ........ 18

Materials and Methods.......................................... 20
Results..........................................................26
Discussion........................................................31

IV QUANTITATIVE APPROACHES TO MEASURING NODULE GAS PERMEABILITY:
DEVELOPMENT OF A RAPID NON-DESTRUCTIVE ASSAY ............... 34

Review of Current Techniques ............................. 35
Development of the Non-Steady-State Lag-Phase Model....... 47
Validation of the Lag-Phase Model with Intact Plant
Material .....................................................65
Conclusions.............................................. 69

V NODULE GAS PERMEABILITY, RESPIRATION AND NITROGEN FIXATION
IN RESPONSE TO ALTERED OXYGEN CONCENTRATION................... 70

Materials and Methods.......................................... 71
Results..........................................................73
Discussion ... ........... ........... .. ...... 79










iii











PAGE

VI SOIL TEMPERATURE, NODULE GAS PERMEABILITY AND DIURNAL CYCLES IN SOYBEAN NITROGEN FIXATION RATE .................... 83

Materials and Methods ....................................... 85
Results ........................................................90
Discussion..................................................... 96

VII CONCLUSION....................................... 100

APPENDIX

BASIC PROGRAM FOR LAG-PHASE SIMULATION...................... 105

REFERENCES................................................... .... 108

BIOGRAPHICAL SKETCH................................ ............... 113











































iv
















Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

REGULATION OF SOYBEAN NODULE GAS PERMEABILITY AND NITROGEN FIXATION RATE

By

PAUL RANDALL WEISZ

December 1986

Chairman: Dr. T. R. Sinclair
Major Department: Agronomy

The gas permeability of leguminous nodules plays an important role in protecting nitrogenase from oxygen inactivation. The purpose of this research was to test the hypothesis that the nodule gas permeability is under active physiological regulation and that changes in the nodule gas permeability can in turn affect nitrogen fixation rates.

The first condition under which this hypothesis was tested was changing rhizosphere oxygen concentrations. Nitrogen fixation rates were immediately affected by alterations in oxygen concentration in a
3 -3 3 -3
range of 0.06 mm mm to 0.40 mm mm These effects were transitory as several hours after altering the oxygen concentration acetylenereduction rates in both intact field- and hydroponically grown soybean plants returned to rates similar to those observed under ambient conditions. Nodule respiration (oxygen uptake) responded to the external oxygen concentration in a similar fashion. The nodule gas permeability, however, varied in response to the alteration in the





V











external oxygen concentration and reached steady-state values significantly different from those observed under ambient conditions.

The second condition under which this hypothesis was tested was

changing soil temperature. Diurnal acetylene-reduction rates of intact field- and hydroponically grown soybean plants were found to be related to the soil temperature even when the diurnal cycles of photosynthetically active radiation and soil temperature were completely out of phase with each other. Changes in soil temperature also resulted in proportional changes in the nodule gas permeability which could not be explained in terms of a passive plant process.

It was concluded that in soybean the nodule gas permeability is a dynamic variable under active physiological control and which in addition to protecting nitrogenase from oxygen inactivation may play an important role in the regulation of symbiotic nitrogen fixation.






























vi















CHAPTER I
INTRODUCTION


Symbiotic nitrogen fixation is a highly energy dependent reaction involving the reduction of gaseous nitrogen to ammonia. The theoretical energy requirement for the reduction of one nitrogen molecule is 12 ATP. Tjepkema and Winship (1980), however, suggested that the actual requirement in soybean (Glycine max) may be at least 24 ATP per nitrogen reduced. This energy requirement is met through nodule respiration which in turn requires carbohydrate and oxygen.

Tjepkema (1971) estimated that in soybean nodules approximately five molecules of oxygen were consumed by bacteroid oxidative phosphorylation for every nitrogen fixed. Therefore, the flow of oxygen into nodules must be considerably larger than that of nitrogen. Nitrogenase, however, is inactivated by even traces of oxygen (Bergersen, 1962; Robson and Postgate, 1980); thus the very system which requires large amounts of energy and, therefore oxygen, must also be protected from oxygen inactivation.

In a review of the function of oxygen in symbiotic nitrogen

fixation Sinclair et al. (1985) suggested that nodule anatomy is one of the main elements of nitrogenase protection from oxygen inactivation. Soybean nodules are approximately spherical and can be divided








2


into three anatomical regions. These regions are the inner most or bacteroid zone where nitrogen reduction takes place, the inner cortex just exterior to the bacteroid zone, and the outer cortex which comprises the outer most tissue (Goodchild, 1977). With the exception of the inner cortex (Tjepkema and Yocum, 1974) most of the nodule volume is permeated with intercellular air spaces (Bergersen and Goodchild, 1973). Due to the presence of these air spaces, oxygen transport through most of the nodule is dominated by the rapid process of diffusion in the gas phase. In the inner cortex however, oxygen transport must take place in the aqueous phase and this has led to the conclusion that this nodule zone acts as a diffusion barrier to the entry of oxygen into the nitrogen fixing tissue in the nodule interior (Tjepkema, 1971; Tjepkema and Yocum, 1974; Sinclair and Goudriaan, 1981). Since the oxygen demand in the bacteroid zone interior to this barrier is high, most of the oxygen crossing the cortical diffusion barrier is consumed by respiration and this results in a very low

-7 3 -3
internal nodule oxygen concentration of approximately 2 10 mm mm (Appleby, 1984). Thus, by placing the site of high respiratory demand and nitrogen fixation interior to a diffusion barrier, the low oxygen environment necessary for nitrogenase to function is established.

Since the thickness of the diffusion barrier in the inner cortex is small compared to the nodule diameter, the flux of oxygen crossing it and entering the bacteroid zone can be modeled as



J = P (Cex in) (1.1)








3


where J is the flux of oxygen entering the nodule and assumed to be approximated by the nodule respiration rate (mm3 mm2 s ), P is the gas
-1
permeability of the diffusion barrier (mm s- ), and O and in are the ex in

oxygen concentrations external and internal to the barrier respectively (mm3 mm-3). Equation 1.1 can be re-arranged to solve for the internal oxygen concentration.



O = O J / P (1.2) in ex


3 -3
Under most conditions O is stable at approximately 0.2 mm mm ex
3 -3
and 0. must be maintained near zero mm mm in order for nitrogenase
in

to function. These two constraints place a major demand on nodule physiology. Maintenance of the low internal nodule oxygen concentration and continued nodule function are dependent on the ratio of the rate of nodule respiration (J) to the nodule permeability (P). For example, if J decreased due to a lack of nodule carbohydrate or environmental stress such as nodule desiccation, Oin would increase and inactivate in

nitrogenase unless there was a subsequent decrease in P. Thus any environmental factor which reduced nodule activity must also result in a reduced nodule gas permeability, or nitrogenase would be inactivated and nitrogen fixation would cease. Likewise, Equation 1.2 indicates that for an increase in J to occur as a result of an elevated nodule photosynthate supply, P must also increase. Failure of P to vary in response to any of these conditions would either result in the inactivation of nitrogenase or in P placing a severe limitation on the nitrogen fixation rate.







4


There is strong evidence that the nodule gas permeability does vary in response to a number of environmental conditions which affect nitrogen fixation rates. Winship and Tjepkema (1983) and Sinclair and Weisz (1985) showed that the nodule gas permeability decreased with decreasing temperature in Alnus rubra and soybean nodules respectively. Weisz et al. (1985) reported that the nodule gas permeability of field grown soybean nodules decreased in response to drought stress and Sheehy et al. (1983) reported that the nodule gas permeability in white clover (Trifolium repens) was responsive to the external oxygen concentration. The purpose of this research was to test the hypothesis that the nodule gas permeability is under active physiological control and that by changing the nodule gas permeablity nitrogen fixation rates may also be regulated.

The first condition under which this hypothesis was tested was altered external oxygen concentrations. Numerous studies (Bergersen, 1962; Pankhurst and Sprent, 1975b; Ralston and Imsande, 1982; Weisz et al., 1985) have indicated that short-term alterations in the external oxygen concentration around nodules result in proportional changes in nodule activity. This is consistent with Equation 1.2 which indicates that if O is increased or decreased, a low value of 0. can be
ex in

maintained by proportional changes in J. There have, however, been a number of studies which indicated that when exposed to long-term alterations in external oxygen concentration there are no concomitant alterations in nodule activity (Criswell et al., 1976,1977; Minchin et al., 1985; Sinclair et al., 1985). In Chapter II these contradicting





4








5


studies are discussed and the long-term effects of altered oxygen concentration on soybean nodule activity are examined. It was found that in the long-term (greater than eight hours) soybean nodule nitrogen fixation rates were independent of the external oxygen concentration in
3 -3
the range of 0.06 to 0.40 mm mm

Equation 1.2 indicates that in order for O. to remain low, when in

0 is altered, J and or P must also change. In Chapter III the nonex
steady-state responses of nitrogen fixation and nodule respiration (J) are examined in response to alterations in 0 Results from these ex

experiments indicate that immediately after altering the external oxygen concentration both the nitrogen fixation rate and the nodule respiration rate change proportionally with the change in 0ex as Equation 1.2 predicts. These altered rates, however, are transitory and within a period of several hours both nodule respiration and nitrogen fixation rates return to values similar to those observed under ambient oxygen conditions. It is proposed that a mechanism exists in nodules which regulates the nodule activity in response to altered oxygen concentration in order to both protect nitrogenase and to maintain a constant nitrogen fixation rate. Furthermore, it is concluded that this mechanism does not involve changes in the nodule respiration rate (J), and it is hypothesized that the adaptive mechanism must involve changes in the nodule gas permeability.

In Chapter IV, the current methods for measuring nodule gas

permeability are discussed. It is found that each of these analytical methods is unsatisfactory and a new procedure is developed. This procedure is then used to measure nodule gas permeability and its








6


response to altered external oxygen concentrations, and the results of these experiments are reported in Chapter V. It is concluded that soybean nodules regulate the gas permeability in response to the rhizosphere oxygen concentration and thus control the rates of nitrogen fixation.

Changing soil temperature was the second condition under which the hypothesis that nodules regulate the gas permeability and are thus capable of controlling nitrogen fixation rates was tested. In Chapter VI, field- and growth-chamber experiments are reported which indicate that diurnal trends in soybean nitrogen fixation rate are the direct result of daily cycles in soil temperature and are independent of diurnal light cycles. These changes in nitrogen fixation rate were associated with proportional changes in the nodule gas permeability and it is suggested that such changes may reflect an active regulation of the gas permeability by the host plant.

The data reported here are consistent with the hypothesis that the nodule gas permeability is under physiological control. Nodule gas permeability was found to be responsive to every environmental stimulus tested which effects nodule activity. Furthermore, these data suggest that the change in nodule gas permeability is not simply a passive response, but one which may be under active regulation. If this is the case nodule gas permeability could play a primary role in both the protection of nitrogenase from oxygen inactivation and in the regulation of nitrogen fixation rates.















CHAPTER II
LONG-TERM EFFECTS OF ALTERED OXYGEN CONCENTRATION ON NITROGEN FIXATION OF INTACT FIELD GROWN SOYBEAN


It has been widely reported that symbiotic nitrogen fixation rates are affected by the oxygen concentration in the rhizosphere around plant roots and nodules. Experiments in which nitrogen fixation rates of detached soybean nodules were assayed as either acetylene-reduction or 15
N2 uptake demonstrated that after exposure to altered oxygen

concentrations nodule activity responded proportionally to the change in oxygen (Bergersen, 1962; Pankhurst and Sprent, 1975b; Ralston and Imsande, 1982). Similar findings have been reported for attached nodule and root systems of Alnus rubra (Winship and Tjepkema, 1983) and in a wide range of legumes including soybean (Glycine max), white clover (Trifolium repens), pea (Pisum sativum), chickpea (Cicer arietinum), cowpea (Vigna unguiculata), peanut (Arachis hypogoea), and lupin (Lupius ablus) (Ralston and Imsande, 1982; Witty et al., 1983; Witty et al., 1984; Winship and Tjepkema, 1985; Weisz et al., 1985).

Respiration as oxygen uptake by both detached nodules and intact

nodulated root systems has also been shown to be sensitive to the oxygen concentration (Tjepkema and Yocum, 1973). Since both nodule respiration and nitrogenase activity are sensitive to the oxygen concentration at which they are assayed, it has been suggested that the oxygen flux crossing a diffusion barrier in the nodule cortex may limit nodule respiration and therefore energy production necessary for nitrogen fixation (Tjepkema, 1971; Tjepkema and Yocum, 1973; Denison et al., 1986).

7








8


All of the experiments cited above were typically completed in less than an hour after the oxygen concentration around the nodules was altered and thus only the short-term effects of altered oxygen concentration were considered. In contrast to these experiments, there are a number of studies which indicate that when exposed to altered rhizosphere oxygen concentrations for long periods of time, nitrogen fixation rates are unaffected by the change in oxygen concentration. These experiments with long-term exposures to elevated or reduced rhizosphere oxygen concentration have failed to show a response in either 1) field-grown soybean nodule activity as measured by acetylenereduction (Criswell et al., 1976, 1977), 2) plant growth and nitrogen content of soybean and pea (Minchin et al., 1985), or 3) plant and nodule mass in field-grown soybean (Sinclair et al., 1985). These data suggest that the role of the rhizosphere oxygen concentration around nodules in limiting nitrogen fixation of intact plants is minimal or non-existent.

Of the three studies on long-term effects of altered oxygen

concentration on nitrogen fixation, Minchin et al. (1985) and Sinclair et al. (1985) inferred fixation rates from growth analysis studies and only Criswell et al. (1976, 1977) assayed nitrogenase activity directly using an acetylene-reduction technique. In the measurements of Criswell et al. (1976, 1977) a non-saturating concentration of acetylene was used. To calculate the expected ethylene production rate at saturating acetylene concentrations (Vmax) these authors used the Michaelis-Menten equation and an assumed Km. This procedure for predicting Vmax failed to take into account the possible effects of diffusion on the











concentration of acetylene at the reaction site inside the nodule (Winship and Tjepkema, 1983; Denison et al., 1983) and therefore may have led to erroneous estimations of nitrogenase activity.

The hypothesis that oxygen limits nitrogen fixation rates in

symbiotic leguminous systems is supported by numerous short-term oxygen studies and conversely is strongly contradicted by the results inferred from two long-term growth studies and experiments possibly using an incorrect estimate of nitrogenase activity. In light of this discrepancy the purpose of this research was to evaluate the long-term effects of altered rhizosphere oxygen concentration on nitrogenase activity of intact field-grown soybean plants.


Materials and Methods


Plant Material And Field Preparation


The cultivar 'Biloxi' was field grown in Gainesville Florida, on Arredondo fine sand (loamy, siliceous, hyperthermic Grossarenic Paleudult). Biloxi is an indeterminate, maturity group VIII cultivar which in this experiment remained vegetative until 14 June 1984 when the first flowers appeared. The experiment was terminated on 19 June, at which time the plants were at growth stage R1 (Fehr et al., 1971).
-I
Field preparation included application of 550 kg ha of 0-10-20

(N-P205-K20) fertilizer and the incorporation of 3 L ha of trifluralin (alpha, alpha, alpha, trifluoro-2, 6-dinitro-N,N-dipropyl-p-toluidine) herbicide. On 2 April, rows spaced 0.9 m apart were seeded with 30 seeds m-1. Open-ended root chambers used for the acetylene-reduction assay (Denison et al., 1983a) were installed in the rows at 1-m








10


intervals immediately after seeding. Alachlor (2-chloro-2',-6'-diethylN-(methyoxymethyl) acetanilide) and chlorpyrifos (0,0-diethyl-0-(3,5,6trichloro-2-pyridyl) phosphorothioate) were then both applied to the plot at a rate of 4 L ha-1. Sprinkler irrigation was applied to the plot to assure well-watered conditions. Acetylene Reduction Assay


In situ ethylene production rates of intact plants in the openended assay chambers were measured at four acetylene concentrations
3 -3
(0.001, 0.004, 0.008, and 0.01 mm mm ) at both mid-day and mid-night over a period of 5 to 7 d. The technique described by Denison et al. (1983b) was used to analyze these data and to determine the mean apparent Km for nitrogenase for all the assays performed (n = 247). The mean and standard error for the Km was found to be 0.00380 and 0.000285
3 -3
mm mm respectively. This mean value was used to calculate Vmax for each assay as described by Denison and Sinclair (1985). Experimental Design


In order to test the long-term effects of altered oxygen

concentration around the nodules, two experimental treatments were used. Each treatment consisting of five plants was assayed repeatedly for 2 to

3 d, and then one treatment was switched from ambient oxygen to either
3 -3
0.06, 0.1, 0.3, 0.35, or 0.4 mm mm in mid-afternoon approximately 8 h before the next acetylene-reduction assay. The other five plants were left at ambient oxygen as a control treatment. Plants in both treatment groups were then further assayed twice a day for 2 to 6 more d. In







11


order to eliminate variability in the Vmax data due to differences among plants in nodule mass, each Vmax estimate was converted to a percentage of the first value observed for a given plant under ambient oxygen conditions. A t-test was used to test for differences between treatment means for normalized Vmax on each day of the experiments.


Results


On 9 June, five plants which had been continuously assayed for 2 d were switched from ambient to 0.06 mm3 mm3 oxygen and assayed for 2 more days. Treatment means for normalized Vmax for this experimental group and a control treatment are presented in Fig. 2.1. After the oxygen concentration was reduced, Vmax appeared to decline; however, this trend started before the change in oxygen concentration occurred and the experimental and control treatment means did not differ significantly (alpha=0.05) on any date before or after 9 June.

On 16 June, plants assayed for 3 d at ambient oxygen concentration

3 -3
were switched to 0.1 mm3 mm oxygen and assayed for another 3 d. Treatment means for normalized Vmax for this experimental group and a control treatment are presented in Fig. 2.2.

Treatment means for Vmax did not differ at the mid-night assay following the afternoon when the oxygen concentration in the experimental chambers was altered. The following noon treatment means

3 -3
appeared to differ with the plants at 0.1 mm mm oxygen having lower acetylene-reduction rates; however, the apparent difference was not statistically significant (alpha=0.05). Treatment mean values for Vmax did not differ significantly on any date of this experiment.






12




1303


120


E
> 110



.c 100

090
O


"q I I I I I
8 9 10 11 12 June
Date







Fig. 2.1. Mean normalized maximum acetylene-reduction (percent of initial value) verses date for two treatments, ambient oxygen
(open circles), and an oxygen regulated treatment switched from ambient to 0.06 mm mm oxygen (solid squares) at the time indicated by the arrow. Treatment means did not differ
significantly (alpha = 0.05) on any date.






13







130

X
0 120
E

- 110


c 100

90



14 16 18 20 June Date






Fig. 2.2. Mean normalized maximum acetylene-reduction (percent of
initial value) verses date for two treatments, ambient oxygen
(open circles), and anoxygn regulated treatment switched from ambient to 0.1 mm mm oxygen (solid squares) at the
time indicated by the arrow. Treatment means did not differ
significantly (alpha = 0.05) on any date.








14


On 17 May, five plants assayed continuously for 2 d were switched from ambient oxygen to 0.3 mm3 mm3. Treatment means for normalized Vmax for this treatment group and a control treatment are presented in Fig. 2.3. Acetylene reduction rates did not differ between treatments over the 24-h period these plants were assayed at 0.3 mm 3 m-3 oxygen. On 18 May, the chamber oxygen concentration was further increased to

3 -3
0.35 mm mm and the plants assayed for another 24 h. On the night of 19 May the experimental treatment mean for Vmax was significantly lower (alpha=0.05) than that of the control plants. Treatment means did not differ significantly on any other date.

On 28 May, five plants assayed for 2 d were switched from ambient

3 -3
to 0.4 mm mm oxygen and assayed for 6 more days. Data from these plants are presented in Fig. 2.4. Treatment means for normalized acetylene-reduction did not differ between oxygen treatments except on
3 -3
one occasion 6 d after the 0.4 mm mm oxygen treatment began.


Discussion


In these experiments acetylene-reduction rates for intact fieldgrown plants were assayed at approximately mid-day and mid-night over a period of approximately 1 wk. At about mid-week, the oxygen concentration in the rhizosphere around five plants was altered about 8 h before the mid-night assay. Thus a response to the altered oxygen concentration which might have occurred in the first 8 h of the treatment would not have been observed. There were no long-term (greater than 8 h) effects of lowering the rhizosphere oxygen concentration to either 0.06 or 0.1 mm3 mm-3 on acetylene-reduction rate.





15




180

x 160

E
> 140

S120
C

o 10080
15 16 17 18 19
May
May Date






Fig. 2.3. Mean normalized maximum acetylene-reduction (percent of
initial value) verses date for two treatments, ambient oxygen
(open circles), and an oxygen regulated treatment switched
3 3 3 -3 from ambient to 0.30 mm mm and 0.35 mm mm oxygen (solid
squares) at the times indicated by the arrows. An "*"
indicates treatment means differ significantly at the alpha =
0.05 level.






16







140



120


-5

-100
0



80



27 29 31 2 4
May DateJune








Fig. 2.4. Mean normalized maximum acetylene-reduction (percent of
initial value) verses date for two treatments, ambient oxygen
(open circles), and an oxyg n regulated treatment switched from ambient to 0.4 mm numm oxygen (solid squares) at the
time indicated by the arrow. An "*" indicates treatment
means differ significantly at the alpha = 0.05 level.








17


Similarly, there were no consistent long-term effects of elevated oxygen concentration. On one date, plants at 0.3 mm mm3 oxygen had significantly lower acetylene-reduction rates but this reduction was not consistent across time. That elevated oxygen concentration did not result in long-term effects on nitrogenase activity is further demonstrated in Fig. 2.4, where plants switched to a rhizosphere oxygen concentration of 0.4 mm3 mm did not differ from the controls in nitrogenase activity over the subsequent 5 d. These data confirm the findings of Criswell et al. (1976, 1977), Minchin et al. (1985) and Sinclair et al. (1985), and further demonstrate that at least after an 8-h of exposure to altered oxygen concentration in the range of 0.06 to
3 -3
0.4 mm mm there are no long-term effects of oxygen concentration on nodule activity.

These data and the data from other long-term oxygen studies

indicate that nitrogen fixation rate in symbiotic leguminous systems is not oxygen limited. Such a conclusion is in contradiction to the numerous short-term studies of nodule activity in many legumes exposed to altered oxygen concentration. The shear number of such short-term experiments and the diversity of experimental procedures used in them make it unlikely that each would contain some common confounding and unknown factor. Criswell et al. (1976) suggested that a mechanism may exist in soybean nodules which allows them to adapt to a wide range of soil oxygen concentration. If such an unknown mechanism exists in nodules and if it took several hours to respond to altered oxygen conditions this could explain the discrepancy between the long- and short-term responses to altered oxygen concentration.















CHAPTER III
NON-STEADY-STATE NODULE RESPIRATION AND NITROGEN
FIXATION IN RESPONSE TO ALTERED OXYGEN CONCENTRATION.


The long-term oxygen studies reported in Chapter II indicated that when the rhizosphere oxygen concentration around intact field grown soybean plants was altered over a range of 0.06 to 0.4 mm3 mm 3, nitrogenase activity was independent of the external oxygen concentration at least after an initial 8-h period. A serious implication of these data on nitrogenase activity at elevated oxygen concentrations concerns nitrogenase protection from oxygen inactivation. Under ambient oxygen the internal oxygen concentration in the infected cells of soybean nodules has been calculated to be on the order of
-7 3 -3
2*10 mm mm (Appleby, 1984). This low internal oxygen concentration is consistent with the fact that nitrogenase is inactivated by oxygen (Robson and Postgate, 1980). Tjepkema and Yocom (1974) demonstrated that the major source of resistance to oxygen diffusion into soybean nodules and the site of the decrease in the oxygen concentration from ambient levels to this extremely low internal partial pressure is a diffusion barrier in the inner cortex of the nodule. Assumming the major resistance to inward diffusion of oxygen is in the nodule cortex, Sheehy et al. (1983) modeled the flux of oxygen into nodules using Fick's first law as



J = P (Oex in) (3.1)




18








19


3 -2 -1
where J is the flux (mm mm s ) of oxygen crossing the diffusion barrier (assumed to be proportional to the nodule respiration rate), P is the gas permeability of the diffusion barrier (mm s-1), 0 is the
3 -3
rhizosphere oxygen concentration (mm mm ) and 0in is the oxygen concentration (mm3 mm 3) in the intercellular air spaces around the bacteroid infected cells inside of the diffusion barrier. As O. is in

many orders of magnitude below 0ex Equation 3.1 can be simplified.



J = P O (3.2) ex



In the experiments described in Chapter II, 0 was increased from
3 -3 3 -3
ambient (0.2 mm mm ) to as high as 0.4 mm mm oxygen without any long-term effects on nodule activity. Equation 3.2 predicts that for a change in 0ex a proportional change in J will also occur unless the permeability of the nodule (P) changes as well. Thus, in order to maintain a low internal nodule oxygen concentration for nitrogenase protection when the rhizosphere oxygen concentration (0 x) is altered, either the nodule respiration rate (J) or the nodule gas permeability

(P) must also change.

Bergersen and Turner (1975, 1980) reported that isolated bacteroids from soybean nodules may contain from two to four different cytochrome oxidase systems which can function at different internal nodule oxygen concentrations. These authors demonstrated that bacteroid respiration rate can vary over a range of dissolved oxygen concentrations from approximately 3*107 mm3 mm3 to 3*104 m mm-3 and yet maintain a fairly constant supply of ATP to nitrogenase for nitrogen fixation.







20


This implies that nodule respiration might vary in response to altered oxygen concentrations and yet nitrogenase activity could remain constant. Such a mechanism is consistent with Equation 3.2 and could explain the results reported for long-term exposures to altered rhizosphere oxygen concentration. A stable low internal nodule oxygen environment would be maintained by an altered respiration rate which would continue to supply nitrogenase with the same amount of ATP resulting in constant acetylene-reduction rates. The purpose of this research was to test the hypothesis that when exposed to long-term alterations in rhizosphere oxygen concentration soybean nodules maintain a low and stable internal oxygen environment by regulating their nodule respiration rates. Also, it is further hypothesized that nitrogen fixation rates as measured by acetylene-reduction are unaffected by changes in oxygen concentration. This predicts that when the external oxygen concentration around soybean nodules is either increased or decreased a proportional change in the oxygen uptake rate will occur without a concomitant change in acetylene-reduction rate.


Materials and Methods


Hydroponically Grown Plant Material


To be consistent with the field experiments reported in Chapter II the cultivar 'Biloxi' was used for these experiments. Seeds were surface sterilized with 2% sodium hypocholorite and germinated on moist filter paper. After germination the seedlings were transferred to growth pouches until the primary root was approximately 50 mm long. The seedlings were then individually transferred to bored #3 rubber stoppers







21


which were placed in the lid of a 1.5 L hydroponic chamber made from 102 mm diameter PVC pipe (Fig. 3.1) and inoculated with a commercial Bradyrhizobium japonicum inoculum (Nitragin Corporation). Plants were maintained in half-strength nitrogen-free nutrient solution (Imsande and
-l
Ralston, 1981) which was continuously aerated by passing 33 mL s of air through an aquarium glass bead bubbler in the bottom of each growth chamber. The growth chambers were submerged in a water bath which regulated the temperature around the roots and nodules to 260C. Illumination was provided by a "Sun-Brella" (Environmental Growth Chambers, Chagrin Falls, Ohio 44022) which consisted of a multi-vapor metal halide lamp (General Electric #E-37) in combination with one highpressure sodium lamp (General Electric #E-18) in a water-cooled jacket

2 -1
and which provided 950 to 1400 uE/m s of photosynthetically active radiation depending on position in the canopy. The photoperiod was adjusted to a 16-h day to assure that plants remained vegetative throughout the assay period.

The day before an individual plant was to be assayed it was

transferred from the hydroponic growth chamber to a stainless-steel flow-through assay chamber (Fig. 3.2). An intact plant, including the rubber stopper through which the stem grew in the hydroponic chamber, was used by inserting the nodulated root system through the bore of the assay chamber until the rubber stopper sealed the top of the chamber (see Fig. 3.2). An air-tight seal at the lower opening of the chamber was made by impregnating the roots in silicon grease (Dow Corning) inside of a split and bored #3 rubber stopper. The assay chamber was then placed on top of a modified hydroponic chamber. Moisturized air













Air
Supply










24 cm

Nutrient Solution



Aerator




Fig. 3.1. Hydroponic growth chamber made from 102 mm diameter PVC pipe
and end cap.








23





















C



D
























Fig. 3.2. Stainless steel flow through assay chamber (A) with
hydroponic support chamber (B). Moisturized gas was supplied
to the assay chamber through port D exited the assay chamber from port C. Air entered the hydroponic chamber through port E. The lower opening of the assay chamber was sealed with a split and bored #3 rubber stopper (G). Roots passing through
the stopper were impregnated in silicon grease (F), and the
stopper was further capped with putty (H).








24



was continuously passed through the assay chamber at a rate of 0.33 mL
-I
s1 to assure that the nodules were well aerated. The assay chamber was maintained in the same water bath and under the same lighting conditions as the growth chambers described above.


Nodule Respiration: Oxygen Uptake


The oxygen concentration in the gas phase was measured with a Walker-type Clark-style oxygen electrode (Delieu and Walker, 1981). Root plus nodule respiration was estimated as the difference between the oxygen concentration in the air supplied to the chamber and that in the gas exiting the chamber. When an experiment was completed, the plant in the assay chamber was removed and the nodules harvested from the root system. The plant was then returned to the assay chamber, and the bare root respiration rate was assayed under the experimental conditions used for that plant. Bare root respiration rate was then subtracted from the root plus nodule respiration rates to yield nodule respiration.


Nitrogenase Activity


Nitrogenase activity of the intact nodules in the assay chamber was determined using the acetylene-reduction technique. To do this the air

-l
supply rate to the assay chamber was increased to 3.33 mL s-1 and

3 -3
acetylene added to a final concentration of 0.10 mm mm To determine the nodule ethylene production rate, 4 min after the acetylene flow was turned on a 1 mL sample was drawn from the gas exiting the assay chamber and injected into a gas chromatograph fitted with a flame ionization detector. The acetylene was then removed from the air supply which was








25


returned to the normal flow rate and any residual acetylene flushed from the chamber.

Saturating concentrations of acetylene have an inhibitory effect on nitrogenase activity in which the ethylene production rate begins to decline after a period of several minutes of exposure (Minchin et al., 1983). This inhibitory effect can cause a serious under estimation of nitrogenase activity depending on the amount of time which has elapsed between when the nodules were initially exposed to the high concentrations of acetylene, and when the ethylene production rate was assayed. Therefore, it was necessary to determine the time dependence of this acetylene response for the hydroponically grown soybeans used in this experiment. To do this acetylene was added to the gas supplied to
3 -3
the assay chamber to a final concentration of 0.10 mm mm and the gas exiting the chamber was sampled every two minutes starting 1 min after the addition of acetylene for a total of 15 min. Nine assays were run in this fashion.


Non-Steady-State Experimental Design


The night before a plant was to be assayed it was transferred from a hydroponic growth chamber to the assay chamber. The following morning nodule activity and respiration were assayed at ambient oxygen. The oxygen concentration in the gas supply was then altered to either 0.1 (for 10 plants) or 0.4 mm3 mm3 (for four plants). The time required for the step change in oxygen concentration to be completed in these experiments was approximately 5 min. Nodule respiration and acetylenereduction were then repeatedly assayed until steady-state was reached.








26


When the experiment was terminated the nodules were harvested and the root respiration assayed at each oxygen concentration used for the individual experiment.


Results


Time Course for Acetylene Inhibition of Nitrogenase Activity


Nitrogenase activity as a percent of the maximum rate of ethylene production observed is plotted against time since the addition of acetylene to the assay chamber for nine individual assays in Fig. 3.3.
3 -3
The concentration of acetylene in the chamber reached 0.10 mm mm after 3 min and maximum ethylene production occurred about 1 min later or a total of 4 min after the addition of acetylene to the gas supply. Acetylene-reduction rates were then stable for approximately 5 more minutes and then began a slow decline until they were 95% of the maximum rate at 14 min.


3 -3
Respiration and Acetylene-Reduction in Response to 0.1 mm mm Oxygen


A total of 10 plants were assayed at ambient oxygen and then
3 -3
switched to 0.1 mm mm oxygen and continually assayed until steadystate was attained. Nodule respiration and maximum acetylene-reduction rate as a percent of that observed under ambient conditions are plotted against time since the growth chamber lights came on in the morning for four representative plants in Fig. 3.4. Nodule respiration rate decreased dramatically when the oxygen concentration was dropped to 0.1

3 -3
mm mm and then recovered slowly over the next few hours. Acetylene reduction rate also decreased when the oxygen concentration was lowered







27








100- ' ' s
0 0
0 0 0 C.









c 40
O






oo


U

E

E


0



0 2 4 6 8 10 12 14 Time (minutes)










Fig. 3.3. Ethylene production rate as a percent of the maximum rate
achieved verses time since the chamber was switched from 0.0
to 0.1 mm3 mm- acetylene.







28







12020


O - sa so

0


S40 40
'Dark Dark .



z P

2





40 40


10 20 30 10 20 30


Time (hours)












Fig. 3.4. Nodule respiration (dots) and maximum acetylene-reduction (open squares) as a percent of the initial value verses time since the growth chamber lights came on for individual plants
assayed under ambient conditions and then switched to 0.1 mm mm oxygen. Arrows indicate the time when the oxygen
concentration was altered.








29


but this decrease was also transitory as there was a recovery to rates similar to those observed under ambient oxygen. A summary of the final steady-state values for nodule respiration and acetylene-reduction rates at 0.1 mm3 mm oxygen is presented in Table 3.1.



3 -3
Table 3.1. Steady-state values at 0.1 mm mm oxygen as
percent initial value at ambient oxygen concentration.


Expt. % Acetylene % Nodule # Reduction Respiration

1 98 121 2 91 80 3 92 104 4 91 5 89 106 6 92 112 7 104 111 8 94 109 9 90 101 10 99 107 MeanS.E: 94.31.7 104.23.6



3 -3
Respiration and Acetylene Reduction Rates in Response to 0.4 mm mm
Oxygen


Nodule respiration and acetylene-reduction were assayed for four
3 -3
plants under ambient oxygen and then switched to 0.4 mm mm oxygen. These data are presented in Fig. 3.5. Nodule respiration decreased after increasing the oxygen concentration around the nodules and then recovered to rates comparable to those initially observed. Non-steadystate acetylene-reduction rates were measured for two of these plants and like nodule respiration initially decreased after increasing the oxygen concentration. Final steady-state acetylene-reduction rates for








30







120 ....20


o 'C

60 o
L
40

00
.I 40 i 140


- 120 20 120






























(open squares) as a percent of the initial value verses time
assayed under ambient conditions and then switched to 0.4
o


10 20 30 10 20 30

Time (hours)














Fig. 3.5. Nodule respiration (dots) and maximum acetylene-reduction
(open squares) as a percent of the initial value verses time since the growth chamber lights came on for individual plants
assayed under ambient conditions and then switched to 0.4
3 -3
mm mm oxygen. Arrows indicate the time when the oxygen
concentration was altered.








31


all four plants were similar to rates initially observed at ambient oxygen. A summary of these steady-state values for nodule respiration and acetylene-reduction at 0.4 mm3 mm3 oxygen is presented in Table

3.2.



3 -3
Table 3.2. Steady-state values at 0.4 mm mm oxygen as
percent initial value at ambient oxygen concentration.


Expt. % Acetylene % Nodule # Reduction Respiration

1 89 96
2 92 99
3 102 98
4 100 90

MeantS.E: 98.83.9 95.02.0



Discussion


Acetylene reduction rates for these intact hydroponically grown soybean plants were assayed four minutes after acetylene was added to the gas supply at which time any residual acetylene was flushed from the assay chamber. This was approximately 5 min before any inhibitory effects of acetylene on nodule activity could be observed (see Fig.

3.3). Minchin et al. (1983) demonstrated that saturating concentrations of acetylene can result in a reduction in nitrogenase activity. The data presented here for soybean, however, indicate that this inhibition 3 -3
was very small being 5% or less over a 14-min exposure to 0.10 mm mm acetylene. For this reason it was concluded that when assayed in the fashion used in these experiments the acetylene-reduction assay did not have an inhibitory effect on nodule activity.








32


As in the field experiments reported in Chapter II, these

hydroponic plants exhibited steady-state acetylene-reduction rates which were independent of the external oxygen concentration from a range of

3 -3
0.1 to 0.4 mm mm3. There were, however, short-term responses of nitrogenase activity to altered oxygen concentrations. Decreasing the
3 -3
oxygen concentration to 0.1 mm mm resulted in decreased ethylene production rates and a dramatic decrease in nodule respiration. These decreased rates were transitory and full recovery of both acetylenereduction and nodule respiration rates were complete in 4 to 8 h. These data indicate that given enough time nodules can adapt to decreased rhizosphere oxygen concentrations. The fact that steady-state nodule
3 -3
respiration rates at 0.1 mm mm oxygen were similar to those observed under ambient conditions does not support the hypothesis that the mechanism for maintaining a constant nitrogen fixation rate under varying oxygen concentrations involves an alteration in nodule respiration rate.
3 -3
Exposure to 0.4 mm mm oxygen resulted in short-term reductions in both nodule respiration and acetylene-reduction rates. Such a response to increased oxygen concentrations may have been caused by partial oxygen inactivation of nitrogenase. The data from the plants
3 -3
assayed at 0.1 mm mm oxygen indicate that the nodule response time to altered oxygen concentrations was several hours. In these experiments the assay chamber oxygen concentration was changed in 5 min. Such a rapid increase in the external oxygen concentration may have overwhelmed the adaptive system. During the period that the oxygen concentration in the chambers was left at these supra-ambient levels, the nodule








33


respiration and acetylene-reduction rates recovered and the final steady-state rates were comparable to those initially observed. Thus, while the external oxygen concentration had been doubled, the nodule respiration rates remained the same. These data are also inconsistent with the hypothesis that the internal low nodule oxygen concentration was maintained by an alteration in nodule respiration rate.

These data confirm that intact soybean nodules have a mechanism for adapting to a wide range of altered rhizosphere oxygen concentrations. This adaption mechanism is time dependent and requires several hours to complete adjustment to the new external oxygen environment. Contrary to the proposed hypothesis, this adaptivity does not involve a change in nodule respiration rate; in fact nodule respiration seems to be as equally affected by this mechanism as does nitrogenase activity.
















CHAPTER IV
QUANTITATIVE APPROACHES TO MEASURING NODULE GAS PERMEABILITY:
DEVELOPMENT OF A RAPID NON-DESTRUCTIVE ASSAY


The data presented in the last two chapters indicate that soybean

nodules contain a mechanism for maintaining a stable low internal oxygen concentration when the nodules are exposed to altered rhizosphere oxygen concentrations. It was hypothesized in Chapter III that this mechanism would involve the regulation of either the nodule respiration rate or the nodule gas permeability. Data presented did not indicate that alteration of nodule respiration was involved in the adaptive mechanism, as nodule respiration and nitrogen fixation rates appeared to respond in similar fashions to altered oxygen concentrations. Since it is likely that the majority of nodule respiration and all of nitrogen fixation take place inside the diffusion barrier, changes in the oxygen flux across this barrier either due to an altered oxygen gradient or to a change in the gas permeability of the barrier could affect both equally. It is hypothesized that the mechanism involved in nodule adaption to altered oxygen concentrations consists of regulation of the gas permeability of the diffusion barrier in the nodule cortex.

In order to test this hypothesis a reliable method for measuring

the nodule gas permeability is essential. Two of the methods currently used for measuring nodule permeability are reviewed here, and it is concluded that each is unsatisfactory. A new method is developed which can be used with intact plants and which results in estimates of both nitrogenase activity and the nodule gas permeability. This technique is 34







35


then used with intact hydroponically grown soybean plants and the resultant estimated values of nodule gas permeability compared with estimates from other sources.


Review of Current Techniques


Nodule Respiration and Fick's First Law Approximation


Sheehy et al. (1983) modeled whole nodule respiration using a

steady-state one-dimensional diffusion equation similar to Equation 3.1 but rearranged to solve for nodule gas permeability such that



P = J / (O O. ) (4.1) ex in


-I
where P is the gas permeability (mm s ) of the diffusion barrier, J is
3 -2 -1
the oxygen flux density crossing the nodule cortex (mm mm s ), O ex
3 -3
is the external oxygen concentration (mm mm ) and 0. is the oxygen in
concentration in the intercellular air spaces internal to the cortical
3 -3
barrier (mm mm- ). Since O is experimentally defined, P can be ex

estimated from Equation 4.1 if the oxygen flux (J) into the nodule interior is known and given certain assumptions about 0. Since 0. is many orders of magnitude lower than O its absolute value is not very ex
3 -3
significant and these authors assumed it to be equal to 0.001 mm nm oxygen.

To estimate J in Equation 4.1, Sheehy et al. (1983) measured nodule carbon dioxide evolution and made certain assumptions to convert this into an oxygen flux. The first assumption concerned the fraction of total respiration representing a respiratory flux across the diffusion








36


barrier. Not all the carbon dioxide evolved by nodules is from the nodule interior, as cells of the inner and outer cortex are also respiring. Quantitative measurements of the respiration rate of the cortical layers have not been made, although Sprent (1972) reported that cells in the nodule cortex show intense metabolic activity. The vascular tissue is also located in the inner cortex, so respiration associated with the loading, unloading and transport of carbon and nitrogen compounds must take place in this region. Witty et al. (1983) estimated the percent of nodule respiration that remained when nitrogenase activity was extrapolated to zero to be about 20% of the total carbon dioxide evolved under ambient conditions. Sheehy et al. (1983) then used this figure to represent the fraction of total respiration taking place in the nodule cortex, or outside the diffusion barrier and thus set J in Equation 4.1 to be equal the measured respiration rate minus this value. This assumes 1) that little or no respiration associated with nitrogen assimilation takes place in the nodule cortex, 2) that respiration associated with cortical or vascular transport is minimal, and 3) that respiration associated with the nodule interior other than that associated with nitrogen fixation is negligible. Errors in any of these assumptions will lead to proportional errors in the resultant value of P from Equation 4.1. Furthermore, should the ratio of cortical to interior respiration vary across experimental treatments, this method of determining P will yield an erroneous change in P.

To estimate J from carbon dioxide evolution rates it is also necessary to know the relationship between oxygen uptake and carbon







37


dioxide evolution. Sheehy et al. (1983) assumed that the respiratory quotient (ratio of carbon dioxide evolved to oxygen uptake) was constant across time and experimental treatments, and equal to unity. Bergersen (1971) reported that the respiratory quotient for soybean nodules decreased from 1.31 to 1.05 as the external oxygen concentration around

3 -3
the nodules was increased from 0.008 to 0.24 mm mm Winship and Tjepkema (1985) also reported that the respiratory quotient for Alnus rubra nodules decreased form 1.2 to 1.0 as the external oxygen

3 -3
concentration was elevated from 0.2 to 0.4 mm mm These reported respiratory quotients are highly variable and are apparently sensitive to at least one of the experimental treatments under which nodule gas permeability has been estimated with this method. For example, Sheehy et al. (1983) used this estimation procedure and reported that P appeared to be responsive to changes in the external oxygen concentration around white clover nodules. When 0 was increased from ex
3 -3
0.21 to 0.8 mm mm the estimated nodule permeability decreased by 76%. If it is assumed however, that the respiratory quotient was decreasing (as Bergersen, and Winship and Tjepkema's data indicate) then as the oxygen concentration around these white clover nodules was raised, the estimated P's would indeed appear to be lower at higher values of 0 ex

The assumption of a constant respiratory quotient is apparently a serious limitation to the use of this model for estimating nodule gas permeability. It was concluded that this analysis procedure is unsatisfactory for testing the hypothesis that nodule gas permeability is involved in the adaptive mechanism to altered rhizosphere oxygen concentrations.








38


Combined Model Using the Diffusion and Michaelis-Menten Equations


Winship and Tjepkema (1983) and Denison et al. (1983b) developed a technique for estimating nodule permeability based on the acetylenereduction assay. The model assumes the reduction of acetylene to ethylene by nitrogenase can be modeled using the Michaelis-Menten equation if the effect of the external diffusion barrier on the concentration of acetylene at the enzyme site is known. A detailed development and analysis of the derivation of this model, has been presented by Johnson and Thornley (1985). While each of these authors used differing algebraic forms of the model one representative equation which is equivalent to those presented in all three of the above citations is



V = (Vmax + KM*k + k*A) / 2


( (Vmax + KM*k + k*A ) 4*k*Vmax*A ) / 2 (4.2) ex ex

3 -1
where V is steady-state nodule ethylene production rate (mm s ), Vmax is the maximum ethylene production rate attained at saturating
3 -1
concentrations of acetylene (mm s ), KM is the apparent nitrogenase Michaelis-Menten constant for acetylene divided by the solubility of
3 -3 3 -1
acetylene (mm mm ), k is the nodule gas conductivity (mm s the product of nodule gas permeability (P) times nodule surface area), and
3 -3
A is the external concentration of acetylene (mm mm ). If the value of V is known at a series of different A 's then the three unknown parameters, Vmax, KM, and k can be solved for using a non-linear regression (Winship and Tjepkema, 1983; Denison et al. 1983b; Denison







39


and Sinclair, 1985). The nodule gas permeability can then be calculated once the total nodule surface area has been estimated such that



P = k / Narea (4.3)



where Narea is the total nodule surface area (mm 2). The nodule surface area can be estimated by the method of Weisz et al. (1985) using measurements of the cross-sectional dimensions of the nodules.

Use of the above approach to determine P is, however, dependent on the use of the acetylene-reduction assay which as described in Chapter III may inhibit nitrogenase activity and thus result in substantial under estimations of Vmax (Minchin et al., 1983; Witty et al., 1984; Minchin et al., 1985). To avoid these acetylene effects, Denison et al.
3 -3
(1983b) used a series of acetylene concentrations below 0.01 mm mm which is an order of magnitude below those at which Minchin et al. (1983) reported the acetylene effects. While this approach avoids the confounding effects of acetylene on the parameters being assayed, it limits the range of acetylene concentrations over which V can be assayed. Thus the regression procedure to solve for the three unknown parameters in Equation 4.2 uses data obtained from only a very small portion of the total acetylene response curve. Furthermore, if only four concentrations of acetylene are used as done by Weisz et al. (1985) only one degree of freedom for the error term in a non-linear regression is left and the ability of the procedure to estimate the parameters with a significant degree of reliability is severely restricted. To improve upon this situation, Denison and Sinclair (1985) assumed that the KM was








40


3 -3
constant and equal to 0.0035 mm3 mm 3 This resulted in a reduced form of the model such that



V = (Vmax + 0.0035*k + k*A) / 2


( (Vmax + 0.0035*k + k*A )2 4*k*Vmax*A ) / 2 (4.4) ex ex


This provided one more degree of freedom for the error term in the regression procedure. If, however, the KM is not constant or the wrong KM value is used in Equation 4.4, eliminating it from the full model may introduce systematic error into the predicted values of Vmax and k.


Sensitivity of the Reduced Model to Errors in the Assumed KM


To test whether Equation 4.4 would predict correct values of k and Vmax if the true KM were indeed variable, a series of nine data sets were simulated. Equation 4.2 was used to generate values of V at four concentrations of acetylene (0.001, 0.004, 0.008, 0.01 mm3 mm 3) using
3 -3
three values of KM (0.002, 0.004, and 0.008 mm mm ) three values of k
3 -1 3 -1 (6.0 12.0 and 24.0 mm s ), and one value of Vmax (0.15 mm s ). These values were similar to those reported for intact soybean nodules (Weisz et al., 1985). Then Equation 4.4 was used with a non-linear regression to solve each of the simulated data sets for a predicted k
3 -3
and Vmax assuming a KM value of 0.004 mm mm The results of this simulation are presented in Table 4.1.








41


Table 4.1. Sensitivity of the reduced model to the assumed KM.

------------------------------------------------------CORRECT K~ CORP CT % ERROR IN THE PREDICTED:
(mm mm ) (mm s ) k Vmax
------------------------------------------------------0.002 6.0 4.7 30.2 0.002 12.0 13.3 22.3 0.002 24.0 42.9 16.7

0.004 6.0 0.0 0.0 0.004 12.0 0.0 0.0 0.004 24.0 0.0 0.0 0.008 6.0 -10.5 -23.7 0.008 12.0 -34.3 -3.7 0.008 24.0 -36.3 -18.5

Percent error in k and Vmax estima es sing an assumed KM of 0.004 mm mm and fixed Vmax of 0.15 mm s


When the reduced model was used to predict k and Vmax from these simulated data sets the values of the predicted parameters were highly accurate when the correct KM was used. When the KM was not constant but
3 -3
varied from the assumed value of 0.004 mm mm the predicted nodule conductances, and maximum rates of ethylene production deviated by -36.3 to 42.9% from the correct values. Large errors can result from the use of this model to predict k and Vmax if the assumed KM is in error, or varies over the experimental treatments being used. This represents a serious limitation to the use of Equation 4.4 to predict nodule conductance.


Reduced Model with Fixed Vmax


In contrast to the approach of Denison and Sinclair (1985) who fixed the value of KM in Equation 4.4 and then solved for k and Vmax, Winship and Tjepkema (1983) measured ethylene production rates of Alnus








42


rubra nodules at a series of acetylene concentrations including saturating levels. The observed value attained at saturating acetylene was assumed to represent Vmax which was substituted into Equation 4.2. A non-linear regression was then used to solve for KM and k. In this manner all three parameters were allowed to vary from plant to plant and across treatments.

The method of Winship and Tjepkema (1983) for determining KM, k and Vmax was tested for soybeans using intact hydroponically grown plants. The plant growth conditions and assay chamber were identical to those described in Chapter III. Acetylene-reduction was assayed at 0.10

3 -3
mm mm acetylene as described in Chapter III, except instead of flushing the chamber with fresh air after exposure to saturating acetylene, the acetylene concentration was altered in order to additionally measure ethylene production at 0.001, 0.005, 0.015 and
3 -3
0.025 mm mm acetylene. The observed ethylene production rate at 0.10

3 -3
mm mm acetylene was assumed to represent Vmax and was substituted into Equation 4.2 which was used with the remaining data to solve for k and KM. A total of 15 plants were assayed in this fashion at ambient

3 -3
oxygen, ten were assayed at 0.1 mm mm oxygen and two were assayed at
3 -3
0.4 mm mm oxygen.

Observed ethylene production rates for a representative plant

assayed at ambient oxygen at four acetylene concentrations are shown in Fig. 4.1. The "best fit" results from the non-linear regression for K4 and k were used to generate the lower line which results in an R2 of better than 0.999. Surprisingly, choosing a value of k which was 50% higher than the one obtained by the non-linear regression could also be used to generate a curve which approximated the actual data if the KM






43







0 10
X k=10.5 mm3/s u) km =0.0083 mm3/mm3
R 2 0.999
E
E 6

.O [k =7.01 m 3/ u 4- km=0.0051 mm3/mm3
:: R2=0.999


1_ 2
\I 1 50% variation in k

0 0.6 1.2 1.8 2.1

C2H2( m m mm3) x 102







Fig. 4.1. Ethylene production verses acetylene concentration for an intact hydroponically grown soybean (dots). Lines represent
curves generated by Equation (4.2) using the correct Vmax, and two different KM's and values of k which differ by 50%.
Error bars represent 95% confidence intervals for the
individual ethylene production rates and acetylene
concentrations.








44


was also increased. This second curve is the upper line in the figure and the resultant R2 for this fit is also better than 0.999. Both of these curves are within the 95% confidence intervals for the individual data points. A family of curves each with different values of k and KM can be fit to the data using Equation 4.2, each of which will closely approximate the data and result in a low error sum of squares. The curve which the non-linear regression predicts as a best fit may, therefore, be more a factor of random or systematic error in the data than an actual indication of the true value for k. This represents a serious limitation to the use of Equation 4.2 for prediction of nodule conductance even if Vmax is known.


Validation of Vmax Estimates for the 1984 Field Data


In Chapter II and VI estimates of Vmax from field studies using the reduced model with an assumed KM are reported. It has been demonstrated that this model can not be used to predict accurately the nodule conductance, k, if the KM is unknown or changing over the experimental treatments. Also the simulations reported in Table 4.1 indicate that depending on the degree of error in the assumed KM and the true value of the nodule conductance, the resultant estimation of Vmax from the reduced model might also be incorrect.

To test the validity of the Vmax estimates from the field studies the data from the 27 assays of intact hydroponically grown plants assayed at 0.1, 0.2 and 0.4 mm3 mm3 oxygen described above were used. The analysis procedure used with the field plants (see Chapter II) was repeated with these data such that Equation 4.2 was used to estimate a








45


mean K fc; the entire data set. This mean KM was then used in Equation

4.4 to estimate the k and Vmax associated with each assay using the ethylene production rates observed at the three lowest acetylene concentrations which were 0.001, 0.005 and 0.015 mm3 mm-3. In this way the estimated Vmax could be compared with the observed maximum rate measured at saturating acetylene concentrations. The results from these assays are reported in Fig. 4.2 where estimated Vmax is plotted against the observed maximum rate. There is a good linear fit (R2 = 0.91) and the slope of the regression is 0.80 which indicates that the estimated values of Vmax are about 20% under estimated. While the absolute value of the estimates is low, the error is consistent across the full range of rates measured. In light of this, Vmax estimates reported in Chapters II and VI were presented as percent of an initial value observed under control conditions. This de-emphasized the absolute value of the estimates and yet allowed comparisons of Vmax across treatments.


Conclusions


In theory Equation 4.2 can be used to predict the Vmax, k, and KM for nitrogen fixing nodules. However, as acetylene may inhibit nitrogenase activity, unless the assay can be completed rapidly it is necessary to use concentrations well below those at which nitrogenase is saturated. This results in data sets which are difficult to resolve into reliable estimates with a non-linear regression. If the KM is assumed to be constant and known, then the model can be used to predict Vmax and k even at these low acetylene concentrations. If however







46








40 1" I CJ


(1)


E30




E

20


U I


L


0 20 30 40

Vmax (mm3/s)xl 02









Fig. 4.2. Estimated values of Vmax verses measured maximum rates of acetylene-reduction gor pEants assayed under ambient and
either 0.1 or 0.4 mm mm oxygen. Estimates were calculated
from eth lene3production rates measured at 0.001, 0.005 and
0.015 mm mm acetylene.








47


the assumed KM is in error or changes during the experiment, large errors in k and Vmax may occur. Finally, the shape of the acetylene response curve for various values of k and KM at a given Vmax reflects an interaction between these parameters, and various values of k and KM can result in similar curves over the concentration range used. Unless the KM is known for each experimental treatment, this model may give erroneous values for k. It is concluded that this model can be used with caution to predict Vmax, but is unreliable for predicting the nodule gas conductance and is therefore unsatisfactory for testing whether the nodule gas permeability adapts to altered rhizosphere oxygen concentrations.


Development of the Non-Steady-State Lag Phase Model


Davis (1984) suggested that the time required for nodules to reach steady-state ethylene production rates after being exposed to acetylene should be a function of both the geometry of the nodules and the diffusivity of the nodule tissue. Davis (1984) measured this "lagphase" for acetylene-reduction in detached nodules from Sesbania rostrata and hairy winter vetch and calculated the time required to reach steady-state ethylene production. To compare reaction times from nodules of different shapes or sizes, Davis (1984) assumed that nodules were homogeneous and nonreactive through out. Both of these assumptions he admitted are incorrect, as the nodule cortex represents the majority of the resistance to inward gas diffusion and the nodule interior reacts with acetylene. Furthermore, a method for relating the time required to achieve steady-state to nodule gas permeability was not presented.








48


Davis (1984) demonstrated that the lag time to steady-state ethylene production rate does vary from species to species and is dependent on the pretreatment of the nodules used, indicating that the lag time may be a valid indicator of the nodule gas permeability. A method is presented here which can be used to 1) measure the lag time to steadystate ethylene production in intact nodules, 2) calculate the time constant for this lag period, and 3) estimate the mean nodule gas permeability of the nodules assayed from this time constant. Physical Description of the Model System


The diffusion characteristics and anatomy of soybean nodules have been previously described in detail (Spent, 1972; Bergersen and Goodchild, 1973; Tjepkema and Yocum, 1974; Goodchild, 1977; Sinclair and Goudriaan, 1981; Selker and Newcomb, 1985; Sinclair et al., 1985). In this model nodules are assumed to be spherical with the majority of the resistance to gaseous diffusion occurring in the nodule cortex which corresponds to an outer layer of the sphere. Internal to this diffusion barrier a system of intercellular air spaces permeates the nodule and therefore, the diffusivity of gases in this region is rapid compared to the diffusion rate through the cytoplasm of the cortical zone. In light of this high rate of internal diffusivity, Sinclair and Goudriaan (1981) predicted that the concentration of gases in the nodule interior would be nearly uniform. It is assumed therefore that the concentration of acetylene and ethylene in the nodule interior is in equilibrium with the inner surface of the diffusion barrier. Once inside the barrier, acetylene reduction to ethylene can be described by Michaelis-Menten kinetics and then the ethylene diffuses back out of the nodule.







49


The nodule system to be assayed is assumed to be intact and in a flow-through chamber. The chamber is also assumed to have perfect mixing and a small time constant (ratio of the gaseous volume of the chamber to the volumetric flow rate through the chamber). If at time zero acetylene is added to the air supply to a final concentration of

3 -3
0.10 mm3 mm there will be a lag period before steady-state ethylene production is reached. This lag should be made up of 5 basic components: 1) the time required for the chamber to reach the final acetylene concentration, 2) the lag required for acetylene to diffuse into the nodules, 3) the time required for the ethylene concentration inside the nodules to reach its final value, 4) the time required for ethylene to diffuse out of the nodules, and 5) the time required for the ethylene concentration in the chamber to reach steady-state. The first and fifth component will depend on the assay chamber volume and the flow rate of the gas supply through the chamber. To minimize the effects of these two components the time constant for the assay chamber used in this analysis and in the following experiments was only two seconds as a result of the rapid volumetric flow rate through an assay chamber of relatively small volume.


Relating Time-To-Steady-State with the Nodule Gas Permeability


Because the time required to reach steady-state is a function of the permeability of the diffusion barrier and the nodule volume, and because the ethylene production rate is governed by Michaelis-Menten kinetics, there is no direct relationship between these parameters which lends itself to an analytical solution. Thus, in order to relate P with







50


the time to steady-state, numerical methods must be employed. To do this a series of nodules were simulated in order to generate a table of predicted times-to-steady-state for nodules of various dimensions and with different values of P.

Simulated nodules were assumed to be spherical and to have outer diameters of 2.0, 2.5 and 3.0 mm. The corresponding diameters of the nitrogen fixing inner zone for each of these nodule sizes were calculated according to Bergersen (1982) as being 1.48, 1.84 and 2.38 mm, respectively. Thicknesses used for the diffusion barrier were 10, 20, 30, 40, 50, 60 and 70 micrometers, and the value of the diffusivity of acetylene and ethylene was calculated from published values for 02 diffusivity in plant tissues (Berry and Norris, 1949; Dungey and Pinfield, 1980) which were extrapolated to 26oC (Carlson, 1911) as
-4 2 -1
4.6*10 mm s The solubility of acetylene was assumed to be 1.C and that of ethylene to be 0.108 (Orcutt and Seevers, 1937). These assumed
-3
values resulted in simulated values of P which ranged from 6.57*10 to
-2 -1
4.60*10 mm s .

If the time constant of the assay chamber is known, and if perfect mixing occurs in the chamber then the acetylene concentration at the outer surface of the nodule can be modeled as



A A (1 et/tau (4.5) ex fin



where Aex is the concentration of acetylene in the assay chamber and at

3 -3
the outer most surface of the diffusion barrier (mm3 mm ), Afin is the fin
final concentration of acetylene in the chamber (mm3 mm-3) after mixing








51


3 -3
has been completed and is assumed to be 0.10 mm mm t is the time elapsed since acetylene was introduced to the chamber (s), and tau is the chamber time constant assumed to be 2 sec.

Since the assumed thickness of the diffusion barrier (10 to 70

micrometers) is small compared to the diameter of the sphere (1.48 tc

2.38 mm), diffusion across the barrier can be modeled with slab geometry. To do this the barrier was divided into compartments 1.25 micrometers in width and the diffusion of acetylene and ethylene across these compartments was modeled with a continuity equation such that



dc/dt = (flow. flow + generation) / vol. (4.6) in out I



where dc/dt is the change in concentration of the diffusing gas with
3 -3 -1
time (mm mm s ), flown and flowout are the flow rate of the gas
3 -1
into and out of the compartment respectively (mm s ), generation is
3 -1
the rate of generation of the gas in the compartment (mm s ), and vol.
I

is the compartmental volume (mm3). Since ethylene is not generated in the diffusion barrier, Equation 4.6 can be used to describe the inward diffusion of acetylene such that



dCa./dt = D (Ca.i+ 2*Ca. + Cai-) / (1.25 10-3 ) (4.7)



3 -3
where Ca. is the concentration of acetylene (mm mm ) in the ith
1

compartment (with i=O at the outer edge of the barrier), and D is the
2 -1
diffusivity (mm s ) of acetylene in the compartment. Once the







52


acetylene reaches the inner surface of the barrier it enters the inner volume of the sphere where it is consumed such that



R = (Vol Vm Can) / (Km + Can) (4.8)



3 -1
where R is the rate of ethylene production (mm s ), Vol is the volume of the sphere inside the diffusion barrier (mm3), Vm is the maximum rate of ethylene production per unit volume (initially assumed to be 6.5*10-4

3 -3 -1
mm3 mm s ), Ca is the concentration of acetylene in the inner most
n
3 -3
compartment of the diffusion barrier (mm mm ), and Km is the Michaelis-Menten constant for reduction of acetylene by nitrogenase (initially assumed to be 0.004 mm3 mm- ).

The outward diffusion of ethylene in each compartment was also modeled using the continuity equation and resulted in a series of equations similar to Equation 4.7. The Euler's finite difference method was used on a mini-computer to solve this series of differential equations and the simulation was run until the calculated rate of ethylene leaving the assay chamber was greater than 88% of the predicted final steady-state value which was assumed to be the value of R when Ca
n
3 -3
was 0.10 mm mm For each nodule size and barrier thickness which was simulated, the calculated rate of ethylene flow out of the assay chamber was plotted as a function of time. Fig. 4.3 is a series of such






53





10

0
x 8



E 6



3 Permeability (mrrV/s)xlO0
-0 46.0 1a 2- o 23.0 S 11.5 a 6.57 U0 20 40 60 80 100

Time(seconds)






Fig. 4.3. Simulated ethylene flow rate out of the assay chamber (Fe)
for a 2.0 mm diameter nodule at four different barrier
permeabilities.







54


simulated data for 2.0 mm outer diameter nodules at four different barrier permeabilities where P is calculated as



P = D S / Lx (4.9)




where S is the solubility of acetylene in the cytoplasm (assumed to be

1.0), and Lx is the thickness of the diffusion barrier (mm).

Except for the the initial few seconds of the simulation each of these curves appear to be exponential in nature and to fit the general exponential equation



-t/tu
F = Ffin (1 e ) (4.10) e fin


3 -1
where F is the ethylene out flow (mm s ) from the chamber at time t
e
3 -1
(s), F in is the final steady-state ethylene out flow rate (nrm s ), and tu is the associated time constant (s). Equation 4.10 can be transformed to a linear equation such that



In(l F /F ) = -(l/tu) t (4.11) e fin



where In is the natural logarithm. Equation 4.11 describes a straight line with slope -(1/tu). Thus a simple linear regression can be used to find the time constant (tu) associated with the simulated data sets for each nodule size and permeability. Figure 4.4 is the same data presented in Fig. 4.3 which has been transformed according to Equation

4.11. The data from each of the simulations was transformed in this





55






Permeability (mr/'s) x 10 3
6.57 0-4- 11.5 o 23.0 a 46.0
0.8
L

Ci 1.2
c

1.6


I I I I I
0 20 40 60 80 100 Ti me (seconds)






Fig. 4.4. Linear transformation of the simulated ethylene flow rate out
of the assay chamber for a 2.0 mm diameter nodule with four
different barrier permeabilities.








56


manner, and a linear regression run on the data which followed the initial lag period or after t = 10 s. Results from this analysis for the simulated data are presented in Table 4.2 where the estimated time constants for steady-state ethylene production for nodules of three diameters and seven nodule gas permeabilities are presented.



Table 4.2. Estimated time constants for simulated nodules ---------------------------------------------------------Barrier Nodule gas Estimated Time Constants (s) Length Permeability For Nodule Diameters (mm) Of
(um) (mm s-1 10 ) 2.0 2.5 3.0
--------------------------------------------------------10 46.0 6.39 7.74 9.73 20 23.0 11.50 14.26 18.36 30 15.3 17.52 21.63 27.86 40 11.5 23.75 29.26 37.59 50 9.20 30.13 37.13 47.55 60 7.67 36.78 45.13 57.75 70 6.57 43.62 53.40 68.23
--------------------------------------------------------In Fig. 4.5 nodule permeability is plotted against the estimated

time constants for each of the three nodule sizes reported in Table 4.2. Ideally ethylene flow rates from an assay chamber with intact nodules in it could be measured as a function of time and these data regressed using Equation 4.11 to solve for an associated time constant. The nodules could then be harvested, measured, and the mean nodule diameter determined. Figure 4.5 could then be used to find the average nodule gas permeability for the assayed nodules by extrapolating between the values for the closest nodule sizes simulated.





57





50
SEANodule Diameter(mm)


o 2.0
E
30


o 20
E
L .

1 10

1 I I I I I
0 10 30 50 70
Time Constant (seconds)






Fig. 4.5. Nodule gas permeability verses the estimated time constants
for simulated nodules of three different diameters.







58


Sensitivity Analysis of the Lag-Phase Model


The data used in Fig. 4.5 were derived from simulated data which 1) assumed only one value of the Km, 2) used only one value of Vm, and 3) assumed that nodules of only one size were present in the chamber and that this nodule size was known. The sensitivity of the lag-phase model to errors in these assumptions was tested. The relative sensitivity of P to one of the factors in the model is defined here as



RS = (P1-P2)/(F1-F2) F1/P1 (4.12)



where RS is the relative sensitivity of P to factor F, and P1 and P2 are the predicted values of the nodule gas permeability when the factor of concern is F1 or F2, respectively. Equation 4.12 was used to test the relative sensitivity of P to changes in the nodule diameter, the Km and

Vm.

Figure 4.5 indicates that for a given time constant, the estimated values of nodule gas permeability vary greatly for nodules of different sizes. Using these data and setting F1 equal to a diameter of 2.5 mm, RS was calculated for a change in diameter of -0.5 and 0.5 mm, at tu values of 14, 26 and 37 s. The relative sensitivity of the model to errors in nodule diameter are given in Table 4.3.







59


Table 4.3. Relative sensitivity of P to nodule diameter.


CHANGE IN DIAMETER TIME CONSTANT (SECONDS)
(mm) 14 26 37

-0.5 0.94 0.89 0.87 0.5 1.19 1.36 1.34



Errors in nodule diameter resulted in substantial errors in the predicted value of P. The relative sensitivities of P to nodule diameter reported in Table 4.3 range from 0.87 to 1.36. This indicates that for given error in nodule diameter, the estimated value of P will also be in error by a factor of 13% less than the error in diameter to 36% greater than the error in diameter.

In order to test the relative sensitivity of P to either Km or Vm, a second set of simulations was run. In these simulations two
3 -3 -4 additional values of Km (0.002 and 0.008 mm mm ) and Vm (2*10 and
-4 3 -3 -1
8*10 mm3 mm s ) were assumed. Results of these simulations are presented in Tables 4.4 and 4.5.



Table 4.4. Relative sensitivity of P to Km


CHAN E IN 3Km TIME CONSTANT (SECONDS) mm mm 14 26 37

-0.002 0.05 0.05 0.05 0.004 0.03 0.03 0.03








60


Table 4.5. Relative sensitivity of P to Vm CHANGE IN VmI TIME CONSTANT (s)
-3
mmm m s 26 37

-4.5*10-4 0.01 0.01

1.5*10-4 0.00 0.01



The relative sensitivities of P to changes in Km and Vm are quite low ranging from 0.03 to 0.05 and 0.0 to 0.01 respectively. This indicates that if either the Km or the Vm of the nodules being assayed differs from the values used to generate Fig. 4.5 by as much as 100%, the error in the predicted nodule gas permeability would be 5% or less.


Predicting P for Systems of Mixed Nodule Sizes


Davis (1984) demonstrated that the time required for a mixture of

nodules of various sizes to reach steady-state ethylene production could be dominated by the larger nodules. The heterogeneous composition of nodule sizes which are found on intact plant roots could therefore introduce an error in the estimations of nodule gas permeability derived by the lag-phase technique. To test whether the lag-phase model could predict the nodule gas permeability of a mixture of nodules of different sizes, simulated data from 2.0, 2.5 and 3.0 mm nodules were combined to form a mixed data set in which there were equal numbers of each of these nodule sizes. This combined data set was transformed using Equation

4.11 and a linear regression used to solve for the associated time constant. The mean nodule diameter for the mixed system was found by calculating the total nodule surface area in the simulated chamber,







61


dividing this by the number of nodules in the chamber to estimate a mean surface area per nodule, and then calculating the spherical diameter which would result in this surface area. The resultant mean nodule diameter for an equal mixture of nodules from each size group was found to be 2.2 mm. This simulation was repeated for three different values of P which were 0.092, 0.0115, and 0.0153 mm s-1. The data in Fig. 4.5 were then used to extrapolate for the estimated P for each of these mixed systems. Table 4.6 gives the resultant percent error in the predicted values of P.



Table 4.6. Percent error in P for a mixed nodule system with
equal numbers of nodules at 2.0, 2.5, and 3.0 mm.

----------------------------------------PERMEABIL{TY PERCENT ERROR IN THE mm s PREDICTED PERMEABILITY
----------------------------------------0.0092 -12.9

0.0115 -13.7

0.0153 -13.7



Mixing an equal number of 2.0, 2.5, and 3.0 mm diameter nodules does introduce an error into the estimated values of gas permeability. The error is approximately -13% and is consistent across the full range of permeabilities tested.

Since it is unlikely that a plant will have an equal number of nodules at each of the sizes that were simulated, nodule size distributions from intact hydroponically grown plants were examined. Nodule data were collected from 19 plants which were used to test the lag-phase technique and which are described below. The nodules were







62


harvested, individually measured and ranked into three size categories based on the average nodule diameter. Simulated data sets consisting of mixtures of nodules of different sizes were then constructed based on the results of these rankings. Most of the 19 plants had nodule distributions dominated by nodule diameters of less than 3 mm, and with an approximate nodule distribution of 48% at 2 mm, 37% at 2.5 mm and 15% at 3.0 mm. A simulated data set with this distribution was considered to represent an average plant. Two simulated data sets were also constructed which represented the most extreme nodule distributions found among the 19 plants that were measured. These extreme data sets represented a plant dominated by large nodules with an approximate distribution of 18% at 2 mm, 29% at 2.5 mm and 53 % at 3 mm, and one plant dominated by small nodules with an approximate distribution of 82% at 2 mm, 16% at 2.5 mm and 2% at 3 mm. Each of these nodule data sets were then simulated with the three values of nodule gas permeability used above. The percent error in the estimated mean nodule gas permeability for each of these mixed nodule data sets are presented in Tables 4.7 through 4.9.



Table 4.7. Percent error in P for a representative mixed
nodule system with 48, 37, and 15% of the nodules at
2.0, 2.5, and 3.0 mm, respectively. PERMEABILITY PERCENT ERROR IN THE
-1
mm s PREDICTED PERMEABILITY

0.0092 9.9

0.0115 9.6

0.0153 -13.1







63


Table 4.8. Percent error in P for a mixed nodule system dominated by large nodules with 18, 29, and 53% of the nodules
at 2.0, 2.5, and 3.0 mm, respectively. PERMEABILITY PERCENT ERROR IN THE
-i
mm s1 PREDICTED PERMEABILITY

0.0092 9.1 0.0115 9.6 0.0153 9.8






Table 4.9. Percent error in P for a mixed nodule system dominated by small nodules with 82, 16, and 2% of the nodules at
2.0, 2.5, and 3.0 mm, respectively.


PERMEABILITY PERCENT ERROR IN THE mm s PREDICTED PERMEABILITY

0.0092 5.0 0.0115 5.2 0.0153 5.2






Each of the simulations of data sets containing nodules of different sizes resulted in underestimations of the nodule gas permeability. In mixtures which contained equal numbers of nodules at each of the three sizes, the error in the predicted value of P was approximately -13%. In the simulated data set which was representative of the plants which are reported below, the error in P ranged from -9.6 to -13.1%. The error associated with P for mixtures which represented plants dominated by large or small nodules was approximately -9.5 and

-5%, respectively.







64


Predicting P for nodules of mixed sizes with the lag-phase

technique resulted in an under-estimation of the correct nodule gas permeability. The magnitude of this error varied from approximately -5% to -14% depending on the nodule size distribution in the mixed sets. This error in the estimated P was relatively small and was very consistent across the range of actual permeabilities simulated for each individual data set. This indicates that while the predicted P for a given plant may be underestimated, this error will be consistent across treatments. Thus the lag-phase technique can be used to determine if the nodule gas permeability of a set of nodules changes over time or in response to an experimental treatment. Conclusions from Simulated Data


The lag-phase method for determining nodule gas permeability

results in estimates which are highly independent of the biochemical characteristics of nitrogenase. This independence from Vm and Km indicate that this method can be used even if these parameters change over the course of an experiment or as part of the experimental treatment. The method is sensitive to errors in the estimate of the mean nodule diameter of the nodules assayed. This problem can, however, be solved as the nodule sizes can be directly measured. Mixtures of differently sized nodules in an assay chamber may cause errors in the absolute value of the predicted nodule gas permeability. This error, however, will be consistent across experimental treatments which might effect the value of the nodule gas permeability. It is concluded that the lag-phase model should predict the nodule gas permeability with a








65


high degree of independence from the biochemical characteristics of nitrogenase activity as long as the mean diameter of the nodules being assayed is known.


Validation of the Lag-Phase Model with Intact Plant Material


Nineteen intact hydroponically grown soybean plants were

individually assayed under ambient oxygen and at 260C to generate data to test the lag-phase model. Plant material, growth conditions, and the assay chamber were identical to those described in Chapter III. The night before a plant was to be assayed it was transferred from a hydroponic-growth chamber to the flow-through assay chamber. The following morning the air flow rate to the chamber was increased from

0.33 to 10.0 mL s-1. This resulted in a time constant for the assay chamber of 2 s. Acetylene was added to the supply gas just upstream
3 -3
from the assay chamber to a final concentration of 0.10 mm mm Onemilliliter gas samples were drawn from the gas exiting the chamber every

4 s until 128 s after the addition of acetylene at which time the acetylene was removed from the gas supply, the chamber flushed with air
-I
and the flow rate returned to 0.33 mL s-1. The gas samples were then injected into a gas chromatograph fitted with a flame ionization detector to determine the ethylene flow rate out of the chamber at the time the individual samples were drawn. The acetylene concentration in the assay chamber and the ethylene production rate reached steady-state in less than 8 s and in approximately 90 s, respectively. Since the assay was complete in less than 3 min, the inhibitory effects of acetylene on nitrogenase activity were avoided (see Fig. 3.3) and the







66


final steady-state rates of acetylene-reduction were assumed to represent maximum nitrogenase activity.

In order to calculate nodule gas permeability the mean nodule diameter for the nodules in the assay chamber was needed. This was calculated by dividing the estimated total nodule surface area in the chamber by the number of nodules to obtain a mean nodule surface area from which an average diameter was calculated as if the nodules were spherical. Total nodule surface area was calculated by the method previously described by Weisz et al. (1985). In brief, 139 nodules were harvested from hydroponically grown plants similar to those used in these experiments. The length (defined as the longest nodule dimension), width and height of each nodule was measured to the nearest half millimeter and the nodule surface area calculated assuming the nodule was ellipsoidal and using a numerical solution to the equation for the surface area of a quadratic surface. These data for nodule surface area were then regressed against the longest nodule dimension. A quadratic model gave the best fit (R2 = 0.98) and is presented in Equation 4.13.



Narea = 3.008 L + 1.210 L2 (4.13)



Narea is the nodule surface area (mm2) and L is the longest nodule dimension (mm). To calculate the total nodule surface area for an experimental plant the nodules were harvested, L was determined for each nodule and Narea calculated according to Equation 4.13. The total nodule surface area assumed to be the sum of the estimated surface areas from each nodule.








67


Once the mean nodule diameter was determined the non-steady-state ethylene production rate data were transformed using Equation 4.11 and a linear regression used to solve for an associated time constant. The data in Fig. 4.5 were then used to extrapolate for the mean nodule gas permeability of the assayed nodules.

Non-steady-state ethylene production data are presented for a

representative plant in Fig. 4.6 where ethylene production as a percent of the steady-state rate and the linearized transformation of these rates are both plotted against time since acetylene was introduced into the gas supply. The resultant R2 for the linearized data was 0.99 and the associated time constant was 24.0 s. The mean nodule diameter found for these nodules was 2.5 mm and the predicted nodule gas permeability
-3 -1
was 13.9*10 mm s

The mean nodule gas permeability for the 19 plants assayed in this
-3 -1
fashion was found to be 13.3*10 mm s with a standard error of the
-3 -1
mean of 0.61*10 mm s These permeabilities are in close agreement with the permeabilities for well-watered field-grown soybeans (5.0*10 3
-3 -1
to 10*10 mm s ) reported by Weisz et al. (1985). Assuming a diffusivity of acetylene through the nodule cortical barrier of 4.6*10
2 -1
mm s as used above, the resultant mean thickness of the diffusion barrier can be calculated to be 30.5 um. This barrier thickness is very similar to that calculated by Sinclair and Goudriaan (1981), who estimated that a barrier of 45 um would be required to reduce the oxygen concentration inside a soybean nodule to values low enough to permit nitrogenase to function. Sheehy et al. (1983) using the respiration model described above reported somewhat lower nodule gas permeabilities






68






100


80- 041
80


0.8
60


"1-12 T S40 C2H4 Production


a Log Transformation
20 6


I I I I I I
0 20 40 60 80 100 120 Time (seconds)








Fig. 4.6. Ethylene production rate as percent of the final steady-state value, and the linear transformation of ethylene production rate verses time in seconds since acetylene was introduced
into the assay chamber.







69


for white clover nodules at ambient oxygen concentration in the range of

1.87*103 to 2.87*10- mm sConclusions


Two methods for estimating the nodule gas permeability were

reviewed and found to be unreliable. A new method based on non-steadystate ethylene production of intact nodules after exposure to acetylene was developed and tested. This technique is rapid, requiring less than three minutes to complete. It gives estimates of both nodule gas permeability and the maximum rate of acetylene-reduction and it is independent of assumptions regarding nodule respiration or the Km for nitrogenase. Finally, the lag-phase model was tested with intact hydroponically grown soybean plants and the resultant values of nodule gas permeability were very close to both theoretical estimates and predictions made using the other techniques. It is concluded that this method for estimating nodule gas permeability is suitable for testing the hypothesis that altered oxygen concentrations result in a regulation in the permeability of the diffusion barrier in the nodule cortex.















CHAPTER V
NODULE GAS PERMEABILITY, RESPIRATION AND NITROGEN FIXATION
IN RESPONSE TO ALTERED OXYGEN CONCENTRATION


In Chapter III it was concluded that in order to maintain a stable and low internal nodule oxygen environment after altering the rhizosphere oxygen concentration the nodule respiration or the gas permeability would have to adjust. The data presented in Chapter III, however, did not support the hypothesis that the adaptive mechanism involved a change in nodule respiration. Sheehy et al. (1983) reported for white clover nodules that the gas permeability of the diffusion barrier night be sensitive to the oxygen gradient across it and vary in magnitude in response to altered external oxygen concentrations. Their hypothesis predicts that altered oxygen concentrations around the nodules will effect the nodule activity, but that the permeability of the nodule cortical diffusion barrier will change over time such that the flux of oxygen into the nodule will return to rates similar to those originally observed under ambient oxygen conditions. This implies that the gas permeability of the nodules may be under physiological control and that by changing this parameter nitrogen fixation rates might be regulated. The purpose of this research was to determine if the nodule gas permeability is responsive to the rhizosphere oxygen concentration, and if so, whether it could be the mechanism responsible for regulating nodule activity in response to alterations in the oxygen environment.







70








71


Materials and Methods


The intact hydroponic soybean plants used in these experiments were grown under the same conditions as described in Chapter III. Individual intact plants were transferred from the hydroponic growth chambers to a flow-through stainless steel assay chamber at the start of an experiment. The mean nodule gas permeability of the nodules in the chamber was assayed using the lag-phase technique described in Chapter IV. To do this the air flow rate through the assay chamber was increased from 0.33 mL s to 10 mL s Acetylene was introduced to the air supply just upstream from the chamber to a final concentration
3 -3
of 0.10 mm mm and l-mL gas samples were drawn from the gas exiting the chamber every 4 s for approximately 128 s. At that time, the acetylene was removed from the gas supply, the chamber flushed with air
-I
and the flow rate returned to the normal rate of 0.33 mL s The gas samples were injected into a gas chromatograph fitted with a flame ionization detector in order to determine the non-steady-state ethylene production rate of the nodules in the chamber at the time the samples were collected. An associated time constant for the non-steady-state response data was determined from which the nodule gas permeability was calculated using Fig. 4.5. The mean nodule diameter of the nodules in the assay chamber was needed for this calculation and was determined after harvesting the nodules at the end of the expermiment.

Mitrogenase activity was assayed as part of the lag-phase technique used for determining the nodule gas permeability. Steady-state acetylene-reduction was typically reached in 60-s to 90-s. Since this assay procedure was completed well in advance of the exposure time








72


nescessary for acetylene to inhibit nodule activity (see Chapter III) the steady-state acetylene-reduction rate was assumed to represent nitrogenase activity.

Nodule respiration rate was also determined on about half the

plants as described in Chapter III. In brief, a Walker-type Clark-style oxygen electrode was used to determine the differance between the oxygen concentration in the gas stream supplied to, and exiting from the assay chamber. This difference in oxygen concentration resulted from root plus nodule respiration. Root respiration rate at ambient oxygen and at the final oxygen concentration to which the plant was exposed, was assayed at the end of each experiment after harvesting the nodules and returning the plant to the assay chamber. Nodule respiration rate was calculated as the difference between total root plus nodule oxygen uptake and bare root respiration rate.


Experimental Design


The night before a plant was to be assayed it was transferred from a hydroponic growth chamber to the assay chamber which was maintained at
-2 -1
260C and provided with 950 to 1400 uE m s of photosynthetically active radiation depending on position in the canopy. The following morning nodule activity and gas permeability were assayed at ambient oxygen. The oxygen concentration in the gas supply was then altered to
3 -3
0.1 mm mm (for 10 plants) and these parameters repeatedly assayed until steady-state was reached. Three plants to act as a control treatment were repeatedly assayed over a 30-h period without switching the oxygen concentration from ambient levels.







73


For exposure to supra-ambient oxygen concentration a 30-L mixing tank was introduced into the air supply line upstream of the assay chamber. This increased the time used for a step change in oxygen concentration from the normal 5 min to over 1 h. This was done in an attempt to avoid nitrogenase oxygen inactivation as may have been observed in the experiments described in Chapter III. Nodule gas permeability and acetylene-reduction for four plants initially assayed



3 -3
assayed. After about 2 h at 0.28 mm mm oxygen the system was S-3
switched to 0.32 mm3 mm-3 oxygen and the plants assayed until steadystate was obtained.


Results


Control Plants


Nodule gas permeability, acetylene-reduction and nodule respiration for three plants were repeatedly assayed under ambient conditions for over 30 h. Data from these control plants as a percent of the initial value are plotted in Fig. 5.1 against time since the growth chamber lights came on the first morning of the experiment. No diurnal trends were apparent in these data as nodule gas permeability, acetylenereduction and nodule respiration were very stable over this 30 h light, dark and light period.


3-3
Study at 0.1 mm3 mm-3 Oxygen


Nodule gas permeability, acetylene-reduction and nodule respiration for two representative plants switched from ambient to 0.1 mm3 mm-3







74










3am 7m 3ar ,m
Respiration D A R K 120 .0150 v max *
1 0 Permeability. ..-....



09 10
0


oo



70
o -120






0 10 20 30 <
Time (hours)










Fig. 5.1. Nodule gas permeability (open circles), maximum acetylenereduction (solid squares), and nodule respiration (dots) as
percent of the initial value verses time since the growth
chamber lights came on for three control plants assayed only
under ambient oxygen conditions.








75


oxygen are presented in Figure 5.2. The effect of reduced oxygen on nodule respiration and acetylene-reduction is similar to that reported for plants in Chapter III. In contrast to these two parameters, nodule gas permeability began to increase when the oxygen concentration was decreased and continued to increase to values well above those observed under ambient conditions.

A summary table of the final steady-state values for nodule gas permeability and acetylene-reduction as percent of the initial values for all plants assayed at 0.1 mm3 mm3 oxygen is presented in Table 5.1. Steady-state rates of acetylene-reduction were similar to those observed at ambient oxygen. Nodule gas permeability however, increased by 63% over the initial values observed under ambient conditions.



3 -3
Table 5.1. Steady state values at 0.1 mm mm oxygen
as percent initial value at ambient oxygen.


Expt. % Acetylene % Nodule Gas # Reduction Permeability

1 89 156 2 119 158 3 110 145 4 107 307 5 109 129 6 111 153 7 106 145 8 95 146 9 116 157 10 104 138 MeanS.E: 106.62.9 163.416.2





76



120" ,
t 160 x 100 --" .

> 80 .140

0 60
o0 120
4 40 T. A



Z 80 D0
a a 0 120


c 60
0 40 B S, 4 100
40
0 8 12 1 32 Time (hours)

Fig. 5.2. Nodule gas permeability (open circles), maximum acetylenereduction (open squares) and nodule respiration (dots) as percent of the initial value verses time since the growth
chamber lights came on for an two plants assayed under
ambient conditions and then switched to 0.1 mm mm-3 oxygen.
Arrows indicate the time when the oxygen concentration was
altered.








77


3 -3
Study at 0.28 and 0.32 mm mm Oxygen


Data for nodule gas permeability, acetylene-reduction and nodule respiration for one of the four plants assayed at supra-ambient oxygen concentrations are presented in Fig. 5.3. Nodule respiration initially decreased with elevated oxygen concentration, then increased to values approximately 20% above ambient rates and finally declined to values similar to those initially observed under ambient conditions. Acetylene reduction rate also decreased with elevated oxygen concentration and then recovered to nearly ambient values. Nodule gas permeability decreased with elevated oxygen concentration and reached final values 34% below the initial levels.

Partial oxygen inactivation of nitrogenase may have occurred in two of the four plants as nodule respiration rate and acetylene-reduction rate initially decreased after elevating the chamber oxygen concentration. The remaining two plants showed little or no short-term effect of oxygen concentration on these parameters. In all four plants elevated oxygen concentration resulted in decreased nodule gas permeability. A summary of the steady-state values of acetylenereduction rate and nodule gas permeability as percent of the initial values for these four plants assayed at supra-ambient oxygen concentration is presented in Table 5.2. Steady-state values of S-3
acetylene-reduction rate at 0.32 mm3 mm-3 oxygen were similar to those observed under ambient oxygen. The mean nodule gas permeability, however, decreased by approximately 32%.







78






x 140 ,, * E 0- 1. Dark 100


120
o
c ..



. .--. ..80 0.


zo 70 00 O (
S60-0


-0d 60

0 10 14 18 22 26 30

Time (hours)










Fig. 5.3. Nodule gas permeability (open circles), maximum acetylenereduction (open squares) and nodule respiration (dots) as percent of the initial value verses time since the growth chamber lights came on for an
individual plant assayed under ambient conditions and then switched to 0.28 and then to 0.32 mm mm oxygen. Arrows indicate the times when
the oxygen concentrations were altered.








79


Table 5.2. Steady state values at supra ambient oxygen
as percent initial value at ambient oxygen.


Expt. % Acetylene % Nodule Gas # Reduction Permeability

1 106 60
2 100 69
3 91 76
4 92 66

MeanS.E: 97.33.5 67.83.3




Discussion


The purpose of this research was to test the hypothesis that the nodule gas permeability is sensitive to the external oxygen concentration and that with alterations in the oxygen environment it adjusts in order to maintain a constant flux of oxygen across the nodule cortical diffusion barrier. This hypothesis predicts that at subambient oxygen concentrations where the oxygen gradient across the nodule cortex is reduced, the nodule gas permeability will increase to allow a greater flux of oxygen into the nodule, and conversely under supra-ambient oxygen concentrations the nodule gas permeability should be decreased. The data on nodule gas permeability from these experiments support this hypothesis. The nodule gas permeability began to increase as soon as the oxygen concentration around the hydroponically grown plants was reduced (Fig. 5.2) and at steady-state had increased approximately 63% over the ambient values. Conversely, nodule gas permeability began to decrease when the oxygen gradient was elevated (Fig. 5.3) and final steady-state values were about 32%








80


decreased over ambient values. Nodule gas permeability did respond to the external oxygen concentration around the nodules of these hydroponically grown soybean plants.

Sheehy et al. (1983) concluded that a similar mechanism for

adjusting the nodule gas permeability to altered oxygen concentrations existed in white clover nodules. Sheehy et al. (1983) reported that the time required for white clover nodules to reach new steady-state respiration (carbon dioxide evolution) rates after an increase in the external oxygen concentration was approximately 45 min. The final rates for respiration and nitrogenase activity reported by these authors was not constant across the range of oxygen concentrations used as
3 -3
respiration at 0.3 mm mm-3 oxygen was 127% higher than that observed at
3 -3
0.05 mm mm oxygen. Witty et al. (1984) reported similar findings for pea and lucerne but were unable to detect an adjustment in nodule gas permeability in soybean or sainfoin. Witty et al. (1985) were also unable to detect an adjustment mechanism in soybean. These findings led Witty et al. (1984) to hypothesis that there are two groups of legumes. In the first group nodules were assumed to be capable of rapid adjustment in nodule gas permeability and this would explain the results of these authors with white clover and pea nodules. The second group of nodules was assumed to have either no, or only a very slow adjustment capability. Soybean was placed in this second group. The experminents of Witty et al. (1984, 1985), however, were only concerned with the short-term response (1 h or less) of the nodules to altered oxygen concentrations and this might explain why a change in nodule gas permeability was not indicated for soybean.








81


The data presented here indicate that soybean nodules are capable

of regulating the nodule gas permeability and that the time required for this regulation is longer than that reported by Sheehy et al. (1983) for white clover. It is possible that the regulator mechanism in soybean is slower than that of white clover, but these data also indicate that it may be more dynamic as complete adjustment of nodule activity was possible over a wide range of external oxygen concentrations (0.1 to 0.4 mm3 mm-3). It is also possible that the legumes assayed by Sheehy et al. (1983) and Witty et al. (1984, 1985) would also have completed full recovery of nodule activity over these oxygen concentrations if the experimental procedures used by these authors had been extended for longer time periods.

Nodules are apparently capable of regulating nitrogen fixation

rates by changing the rate at which oxygen can diffuse into the nodule interior. Oxygen diffusion into soybean nodules is initially dominated by diffusion in air spaces in lenticel like structures in the outer cortex (Pankhurst and Sprent 1975a). Interior to the outer cortex there is a layer of cells in the inner cortex which lack intercellular air spaces and which has been hypothesized to be the site of the oxygen diffusion barrier (Tjepkema and Yocum, 1974; Sinclair and Goudriaan, 1981). Pankhurst and Sprent (1975a) reported that when subjected to drought-stress the air spaces in the lenticels close as the cells in the outer cortex lose turgor, and this led Pankhurst and Sprent (1975b) to hypothesize that a decrease in the nodule gas permeability would be associated with drought. This hypothesis was supported by the data of Weisz et al. (1985) which indicated that drought-stressed soybean








82


nodules had significantly lower values of nodule gas permeability than did nodules on well-watered control plants. Turgor pressure in the lenticular cells seemingly affects the air spaces in the outer cortex and the magnitude of the nodule gas permeability. It is possible that unknown mechanisms which regulate turgor in the cortical cells represent the basis of the nodule capacity for regulating the nodule gas permeability and subsequently nitrogen fixation.














CHAPTER VI
SOIL TEMPERATURE, NODULE GAS PERMEABILITY AND DIURNAL
CYCLES IN SOYBEAN NITROGEN FIXATION RATE


Recent research has indicated that diurnal cycles in symbiotic

nitrogen fixation rate in intact leguminous plants may be the result of temperature changes around the plant roots and nodules. Eckart and Raguse (1980) reported that diurnal cycles in nodule acetylene-reduction rates in growth chamber grown subterranean clover (Trifolium subterraneum L.) were related to the soil temperature. Similarly, Schweitzer and Harper (1980) reported that when maintained at constant root temperature diurnal light cycles did not effect the nodule activity of growth cabinet grown soybean plants (Glycine max L.). Diurnal fluctuations in nodule activity were only evident when the temperature around these roots and nodules was altered. Winship and Tjepkema (1983, 1985) also found that acetylene-reduction rates for intact green housegrown Alnus rubra root nodules were related to nodule temperature.

Further evidence that diurnal cycles in nitrogen fixation rate are related to changes in soil temperature has been reported in studies using intact field-grown soybean. Denison and Sinclair (1985) found that there was a high correlation between in-situ soybean nodule activity and soil temperature measured at a depth of 0.1 m. Similarly, Sinclair and Weisz (1985) found that there was a linear relationship between soil temperature and acetylene-reduction by intact field-grown soybean nodules at temperatures below 300C. In both these studies diurnal variability in nitrogen fixation rate was correlated with changing soil temperature.

83








84


While there is evidence which indicates that diurnal cycles in soil temperature are the environmental cause of daily fluctuations in nodule activity, it is still frequently assumed that diurnal variability in nitrogen fixation is the result of changing levels of photosynthetically active radiation coupled with a rate limiting nodule photosynthate supply (Hardy and Havelka, 1976). The studies of Denison and Sinclair (1985) and Sinclair and Weisz (1985) with field-grown soybean do not support this hypothesis, but their results are based on correlational data and the natural diurnal cycles of photosynthetically active radiation and soil temperature were not experimentally manipulated in a manner which would clearly separate their individual effects. Root temperature and photosynthetically active radiation were experimentally manipulated in the experiments cited above by Eckart and Raguse (1980) and Schweitzer and Harper (1980) but as growth chamber plant materials were used their results may not reflect mechanisms which function under field conditions.

The initial purpose of this research was to confirm that under

field conditions diurnal cycles in soybean nitrogen fixation rate follow the soil temperature and are independent of daily light cycles. To do this, the effects of the light and soil temperature cycles on nitrogen fixation rate were separated by manipulating the phase and amplitude of the temperature diurnal. The results indicated that both the phase and magnitude of cycles in daily soybean nitrogen fixation were independent of fluctuations in photosynthetically active radiation but strictly followed the soil temperature.








85


Denison et al. (in press) postulated that in leguminous systems the flux of oxygen diffusing into nodules acts as the primary limitation to nodule respiration and therefore energy production necessary for nitrogen fixation. This hypothesis predicts that in order for fixation rates to increase in response to elevated temperature the flux of oxygen into the nodules must also be responsive to nodule temperature. Winship and Tjepkema (1983) and Sinclair and Weisz (1985) reported that the nodule gas permeability changed proportionally with temperature in nodules from Alnus rubra and soybean, respectively. Such a change in the nodule gas permeability would affect the rate of oxygen diffusion into the nodule and might explain the elevated rates of nitrogen fixation observed at increasing temperature. The analytical procedures used by Winship and Tjepkema (1983) and Sinclair and Weisz (1985), however, were criticized in Chapter IV where it was demonstrated that the resultant values of nodule permeability derived using these methods may not be reliable. Therefore, the second objective of this research was to use the lag-phase technique described in Chapter IV to determine if the gas permeability of soybean nodules is responsive to temperature and may act to regulate nitrogen fixation rates in response to diurnal cycles in soil temperature.


Materials and Methods


Field Study with Intact Soybean Plants


The cultivar 'Biloxi' was grown in Gainesville, Florida, on Arredondo fine sand (loamy, siliceous, hyperthermic Grossarenic Paleudult). Biloxi is an indeterminate, maturity group VIII cultivar








86


which was at growth stage R1 (Fehr et al., 1971) when this experiment was conducted from 27 June to 3 July, 1984.
-I
Field preparation included application of 550 kg ha-1 of 0-10-20
-I
(N-P205-K20) fertilizer and the incorporation of 3 L ha-1 of trifluralin (alpha, alpha, alpha, trifluoro-2,6-dinitro-N,N-dipropyl-p-toluidine) herbicide. On 2 April, rows spaced 0.9 m apart were seeded with 30

m-l
seeds m-1. Open-ended root chambers used for the acetylene-reduction assay (Denison et al., 1983a) were installed in the rows at 1-m intervals immediately after seeding. Alachlor (2-chloro-2',-6'-diethylN-(methyoxymethyl) acetanilide) and chlorpyrifos (0,0-diethyl-O-(3,5,Etrichloro-2-pyridyl) phosphorothioate) were then both applied to the
-i
plot at a rate of 4 L ha-1. Sprinkler irrigation was applied to the plot to assure well-watered conditions. The crop canopy was closed by the time this experiment was conducted.

The assay chambers were modified to allow heating or cooling of the soil inside the chambers. This was done by wrapping a 2-m length of 10mm diameter copper tubing around the chambers before they were buried in the field. During the experiment, water from a controlled water bath was circulated through the copper tubing to regulate the chamber temperature. The chamber temperature was measured with a thermistor (#44032, Omega Engineering Inc.) that had been potted inside a 2-mm diameter aluminum tube and buried at a depth of 70 mm near the center of the chamber. The electrical resistance of the thermistor was monitored by a computer which also regulated the temperature of the circulating water.







87


The temperature of the soil around four plants was not altered in order to study acetylene-reduction rates under ambient temperature conditions. A temperature-regulated treatment, composed of six plants, was imposed by altering the diurnal soil temperature cycles. To do this the computer controlling the chamber temperature was programmed to expose the plants to a sinusoidal diurnal variation in soil temperature from 19 to 290C. From 27 to 29 June, this experimental temperature cycle was in phase with the daily light cycle with maximum and minimum temperatures being reached at noon and mid-night, respectively. In the evening of 29 June, the phase of the temperature cycle was shifted by 12 h such that the soil temperature in the chambers was 290C at mid-night and 190C at mid-day.

In situ ethylene production rates of intact plants in the openended assay chambers were measured at four acetylene concentrations
3 -3
(0.001, 0.004, 0.008, and 0.01 mm mm ) at 0600, 1200, 1800 and 2400 EST each day throughout the experiment. The analytical technique described by Denison and Sinclair (1985) was used to calculate the maximum rate of acetylene-reduction at saturating acetylene concentrations. To minimize variability among plants which differed in nodule mass, acetylene-reduction rates were converted to a percentage of the rate observed for a given plant at 1800 h on the second day of the experiment.


Hydroponic Study with Intact Soybean Plants


To be consistent with the field experiments the cultivar 'Biloxi' was used for these hydroponic studies. Seeds were surfaced sterilized







88


with 2% sodium hypocholorite and germinated on moist filter paper. After germination the seedlings were transferred to growth pouches until the primary root was approximately 50 mm long. The seedlings were then individually transferred to bored #3 rubber stoppers which were placed in the lid of a 1.5 L hydroponic chamber made from 102-mm diameter PVC pipe (described previously in Chapter III) and inoculated with a commercial Bradyrhizobium japonicum inoculum (Nitragin Corporation). Plants were maintained in half-strength nitrogen-free nutrient solution (Imsande and Ralston, 1981) which was continuously aerated by passing 33.3 mL s-1 of air through an aquarium glass bead bubbler in the bottom of each growth chamber. The growth chambers were submerged in a water bath which maintained the temperature around the roots and nodules at 260C. Illumination was provided by a "Sun-Brella" (Environmental Growth Chambers, Chagrin Falls, Ohio 44022) which consisted of a multi-vapor metal halide lamp (General Electric #E-37) in combination with one highpressure sodium lamp (General Electric #E-18) in a water-cooled jacket
-2I
and which provided 950 to 1400 uE m-2 s of photosynthetically active radiation depending on position in the canopy. The photoperiod was adjusted to a 16-h day to assure that plants remained vegatetive through-out the assay period.

The day before an individual plant was to be assayed it was

transferred from the hydroponic growth chamber to a stainless steel flow-through assay chamber (previously described in Chapter III). An intact plant, including the rubber stopper through which the stem grew in the hydroponic chamber, was used by inserting the nodulated root system through the bore of the assay chamber until the rubber stopper








89


sealed the top of the chamber. An air tight seal at the lower opening of the chamber was made by impregnating the roots in silicon grease (Dow Corning) inside of a split and bored #3 rubber stopper. The assay chamber was then placed on top of a modified hydroponic chamber. Moisturized air was continuously passed through the assay chamber at a
-I
rate of 0.33 mL s-1 to assure that the nodules were well aerated. The assay chamber was maintained in the same water bath and under the same lighting conditions as the hydroponic chambers described above.

Acetylene reduction and nodule gas permeability for the intact nodules in the assay chamber were assayed as previously described in Chapter IV. In brief, the air flow rate to the assay chamber was
-i
increased from 0.33 to 10 mL s Acetylene was added to the gas supply just upstream from the assay chamber to a final concentration of 0.10

3 -3
mm mm O. ne-milliliter gas samples were collected from the gas exiting the chamber every 4 s until 128 s after the addition of acetylene at which time the acetylene was removed from the gas supply,
-I
the chamber flushed with air and the flow rate returned to 0.33 mL s The gas samples were then injected into a gas chromatograph fitted with a flame ionization detector to determine the ethylene flow rate out of the chamber at the time the individual samples were drawn. The final steady-state rates of acetylene-reduction (typically reached in 60 to 90 s after the addition of acetylene to the gas supply) were assumed to represent maximum nitrogenase activity. The non-steady-state ethylene production rate data were then analyzed using the lag-phase technique described in Chapter IV to calculate the nodule gas permeability for the nodules in the assay chamber.







90


The night before a plant was to be assayed, the temperature of the water bath was increased to 280C. The following morning at 0900 EST, acetylene-reduction rate and nodule gas permeability were assayed and the temperature of the water bath was lowered to 240C. The change in temperature took about 2 h and at noon the assay was repeated. The temperature was then lowered a second time to 200C and acetylenereduction and nodule gas permeability assayed again at 1500 EST. This was repeated for a total of four plants. A control treatment consisting of three plants (previously described in Chapter V) was assayed at a constant root temperature of 260C over a 30 h light, dark and light period.


Results


Diurnal Field Studies


Soil temperature of the ambient and temperature-regulated plants is plotted against time in Fig. 6.1A. Ambient diurnal variability in soil temperature was small, being approximately 20C or less. The gradual and slight decrease in mean ambient soil temperature from day 2 through 4 may have been the result of overcast and rainy weather over that 48-h period. Diurnal variation of soil temperature in the temperatureregulated chambers averaged just under 100C, ranging from approximately 19 to 290C. On the evening of the third day the temperature of these chambers was phase shifted by 12 h.

The mean acetylene-reduction rates for each treatment as a percent of that observed at 1800 h on the second day of the experiment are presented in Fig. 6.1B. On each day of the experiment except day four,





91
3 0 .. .. b,: ...

S28 + 26 S24
E
_ 22 0 20
A
130 9
X
E120
-0 110

100
E
C 90 z 80
B
1 2 3 4 5 6 Day
Fig. 6.1. A) Chamber temperature verses day for control (0), and temperature-regulated chambers (solid squares). B) Treatment mean acetylene-reduction rates as a percent of those observed
at 1800 EST on day 2 verses time since the start of the
experiment.








92


the control plants at ambient soil temperature displayed very slight diurnal variability in acetylene-reduction rates with the daily minimum rate being 90% or more of the diurnal maximum. There was a slight decrease in these ambient daily rates from day 2 through day 4. These diurnal cycles in acetylene-reduction rate were in phase with both the soil temperature and the daily light cycles.

In contrast to the plants at ambient soil temperature, the

temperature-regulated plants displayed large diurnal fluctuations in acetylene-reduction rate. These diurnal trends changed phase by 12 h when the temperature diurnal in the chambers was shifted on day 3. Acetylene reduction rates for individual plants in the temperatureregulated treatment for the 24-h period previous to and immediately after the temperature diurnal was shifted on day 3 are plotted against soil temperature in Fig. 6.2. There is a good linear fit, with R2 equal to 0.85.


Temperature and Light Study with Hydroponic Plants


Plants assayed at a constant temperature of 260C over a 30-h light, dark and light period did not have diurnal fluctuations in nodule gas permeability or acetylene-reduction rate (see Fig. 5.1). Acetylenereduction rate and nodule gas permeability are plotted against chamber temperature for each plant assayed in Figs. 6.3A and Figure 6.3B, respectively. Under constant light conditions acetylene-reduction rate and nodule gas permeability decreased as the temperature around the nodules was decreased from 28 to 20'C. Acetylene-reduction rate is plotted against nodule gas permeability in Fig. 6.4. There was a close






93







140

O

E120- (0
>
I00
N100 1 0


E
1-80- % o .

60


20 24 28
Temperature (OC)








Fig. 6.2. Normalized acetylene-reduction rate verses soil temperature for individual plants in the temperature-regulated treatment
from 1800 EST on day 1 until 1800 EST on day 3.







94
o'

7 11




E 9- 8 o
C3\






I I I
O



C O












n11
00


09
L



20 24 28
Temperature (oC)


Fig. 6.3. A) Acetylene-reduction rate verses chamber temperature
(circles) for four individual hydroponically grown plants assayed at three temperatures. B) Nodule gas permeability
verses temperature (solid squares) for the same
hydroponically grown plants assayed at three temperatures.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EX80M1I93_7JWW26 INGEST_TIME 2015-01-21T19:12:42Z PACKAGE AA00026618_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

REGULATION OF SOYBEAN NODULE GAS PERMEABILITY AND NITROGEN FIXATION RATE By PAUL RANDALL WEISZ A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1986

PAGE 2

ACKNOWLEDGMENTS I would like to thank Dr. Thomas Sinclair for allowing me to work on this project and to undertake this research. His encouragement and tenacious optimism have been a foundation of support without which this work would not have been accomplished. I would also like to thank Mr. George Drake for his technical advice and dedicated assistance with the instrumentation and computerized equipment. I would also like to thank my wife for keeping me sane in those bleak moments during this research when the data appeared hopelessly unintelligible. ii

PAGE 3

TABLE OF CONTENTS PAGE ACKNOWLEDGMENTS i i ABSTRACT v CHAPTERS I INTRODUCTION 1 II LONG-TERM EFFECTS OF ALTERED OXYGEN CONCENTRATION ON NITROGEN FIXATION OF INTACT FIELD-GROWN SOYBEAN 7 Materials and Methods 9 Results 11 Discussion 14 III NON-STEADY-STATE NODULE RESPIRATION AND NITROGEN FIXATION IN RESPONSE TO ALTERED OXYGEN CONCENTRATION 18 Materials and Methods 20 Results 26 Discussion 31 IV QUANTITATIVE APPROACHES TO MEASURING NODULE GAS PERMEABILITY: DEVELOPMENT OF A RAPID NON-DESTRUCTIVE ASSAY 34 Review of Current Techniques 35 Development of the Non-Steady-State Lag-Phase Model 47 Validation of the Lag-Phase Model with Intact Plant Material 65 Conclusions 69 V NODULE GAS PERMEABILITY, RESPIRATION AND NITROGEN FIXATION IN RESPONSE TO ALTERED OXYGEN CONCEl^ITRATION 70 Materials and Methods 71 Results 73 Discussion 79 iii

PAGE 4

PAGE VI SOIL TEMPERATURE, NODULE GAS PERMEABILITY AND DIURNAL CYCLES IN SOYBEAN NITROGEN FIXATION RATE 83 Materials and Methods 85 Results 90 Discussion 96 VII CONCLUSION 100 APPENDIX BASIC PROGRAM FOR LAG-PHASE SIMULATION 105 REFERENCES 108 BIOGRAPHICAL SKETCH 113 iv

PAGE 5

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy REGULATION OF SOYBEAN NODULE GAS PERMEABILITY AND NITROGEN FIXATION RATE By PAUL RANDALL WEISZ December 1986 Chairman: Dr. T. R. Sinclair Major Department: Agronomy The gas permeability of leguminous nodules plays an important role in protecting nitrogenase from oxygen inactivation The purpose of thi research was to test the hypothesis that the nodule gas permeability is under active physiological regulation and that changes in the nodule ga permeability can in turn affect nitrogen fixation rates. The first condition under which this hypothesis was tested was changing rhizosphere oxygen concentrations. Nitrogen fixation rates were immediately affected by alterations in oxygen concentration in a 3-3 3-3 range of 0,06 mm mm to 0.40 mm mm These effects were transitory as several hours after altering the oxygen concentration acetylenereduction rates in both intact fieldand hydroponically grown soybean plants returned to rates similar to those observed under ambient conditions. Nodule respiration (oxygen uptake) responded to the external oxygen concentration in a similar fashion. The nodule gas permeability, however, varied in response to the alteration in the

PAGE 6

external oxygen concentration and reached steady-state values significantly different from those observed under ambient conditions. The second condition under which this hypothesis was tested was changing soil temperature. Diurnal acetylene-reduction rates of intact fieldand hydroponically grown soybean plants were found to be related to the soil temperature even when the diurnal cycles of photosynthetically active radiation and soil temperature were completely out of phase with each other. Changes in soil temperature also resulted in proportional changes in the nodule gas permeability which could not be explained in terms of a passive plant process. It was concluded that in soybean the nodule gas permeability is a dynamic variable under active physiological control and which in addition to protecting nitrogenase from oxygen inactivation may play an important role in the regulation of symbiotic nitrogen fixation. vi

PAGE 7

CHAPTER I INTRODUCTION Symbiotic nitrogen fixation is a highly energy dependent reaction involving the reduction of gaseous nitrogen to ammonia. The theoretical energy requirement for the reduction of one nitrogen molecule is 12 ATP. Tjepkema and Winship (1980), however, suggested that the actual requirement in soybean ( Glycine max) may be at least 24 ATP per nitrogen reduced. This energy requirement is met through nodule respiration which in turn requires carbohydrate and oxygen. Tjepkema (1971) estimated that in soybean nodules approximately five molecules of oxygen were consumed by bacteroid oxidative phosphorylation for every nitrogen fixed. Therefore, the flow of oxygen into nodules must be considerably larger than that of nitrogen. Nitrogenase, however, is inactivated by even traces of oxygen (Bergersen, 1962; Robson and Postgate, 1980); thus the very system which requires large amounts of energy and, therefore oxygen, must also be protected from oxygen inactivation In a review of the function of oxygen in symbiotic nitrogen fixation Sinclair et al. (1985) suggested that nodule anatomy is one of the main elements of nitrogenase protection from oxygen inactivation. Soybean nodules are approximately spherical and can be divided 1

PAGE 8

2 into three anatomical regions. These regions are the inner most or bacteroid zone where nitrogen reduction takes place, the inner cortex just exterior to the bacteroid zone, and the outer cortex which comprises the outer most tissue (Goodchild, 1977). With the exception of the inner cortex (Tjepkema and Yocum, 1974) most of the nodule volume is permeated with intercellular air spaces (Bergersen and Goodchild, 1973). Due to the presence of these air spaces, oxygen transport through most of the nodule is dominated by the rapid process of diffusion in the gas phase. In the inner cortex however, oxygen transport must take place in the aqueous phase and this has led to the conclusion that this nodule zone acts as a diffusion barrier to the entry of oxygen into the nitrogen fixing tissue in the nodule interior (Tjepkema, 1971; Tjepkema and Yocum, 1974; Sinclair and Goudriaan, 1981). Since the oxygen demand in the bacteroid zone interior to this barrier is high, most of the oxygen crossing the cortical diffusion barrier is consumed by respiration and this results in a very low -7 3 -3 internal nodule oxygen concentration of approximately 2 10 mm mm (Appleby, 1984). Thus, by placing the site of high respiratory demand and nitrogen fixation interior to a diffusion barrier, the low oxygen environment necessary for nitrogenase to function is established. Since the thickness of the diffusion barrier in the inner cortex is small compared to the nodule diameter, the flux of oxygen crossing it and entering the bacteroid zone can be modeled as J = P (0 0. ) ex in (1.1)

PAGE 9

3 where J is the flux of oxygen entering the nodule and assumed to be 3 -2 -1 approximated by the nodule respiration rate (mm mm s ) P is the gas permeability of the diffusion barrier (mm s ''"), and 0^^ and O^^ are the oxygen concentrations external and internal to the barrier respectively (mm mm ). Equation 1.1 can be re-arranged to solve for the internal oxygen concentration O. =0 J / P (1.2) m ex 3 -3 Under most conditions 0 is stable at approximately 0.2 mm mm and 0. must be maintained near zero mm"^ mm ^ in order for nitrogenase in to function. These two constraints place a major demand on nodule physiology. Maintenance of the low internal nodule oxygen concentration and continued nodule function are dependent on the ratio of the rate of nodule respiration (J) to the nodule permeability (P) For example, if J decreased due to a lack of nodule carbohydrate or environmental stress such as nodule desiccation, 0. would increase and inactivate in nitrogenase unless there was a subsequent decrease in P. Thus any environmental factor which reduced nodule activity must also result in a reduced nodule gas permeability, or nitrogenase would be inactivated and nitrogen fixation would cease. Likewise, Equation 1.2 indicates that for an increase in J to occur as a result of an elevated nodule photosynthate supply, P must also increase. Failure of P to vary in response to any of these conditions would either result in the inactivation of nitrogenase or in P placing a severe limitation on the nitrogen fixation rate.

PAGE 10

4 There is strong evidence that the nodule gas permeability does vary in response to a number of environmental conditions which affect nitrogen fixation rates. Winship and Tjepkema (1983) and Sinclair and Weisz (1985) showed that the nodule gas permeability decreased with decreasing temperature in Alnus rubra and soybean nodules respectively. Weisz et al (1985) reported that the nodule gas permeability of field grown soybean nodules decreased in response to drought stress and Sheehy et al. (1983) reported that the nodule gas permeability in white clover ( Trifolium repens ) was responsive to the external oxygen concentration. The purpose of this research was to test the hypothesis that the nodule gas permeability is under active physiological control and that by changing the nodule gas permeablity nitrogen fixation rates may also be regulated. The first condition under which this hypothesis was tested was altered external oxygen concentrations. Numerous studies (Bergersen, 1962; Pankhurst and Sprent, 1975b; Ralston and Imsande, 1982; Weisz et al., 1985) have indicated that short-term alterations in the external oxygen concentration around nodules result in proportional changes in nodule activity. This is consistent with Equation 1.2 which indicates that if O is increased or decreased, a low value of 0. can be ex m maintained by proportional changes in J. There have, however, been a number of studies which indicated that when exposed to long-term alterations in external oxygen concentration there are no concomitant alterations in nodule activity (Criswell et al 1976,1977; Minchin et al., 1985; Sinclair et al 1985). In Chapter II these contradicting 4

PAGE 11

5 studies are discussed and the long-term effects of altered oxygen concentration on soybean nodule activity are examined. It was found that in the long-term (greater than eight hours) soybean nodule nitrogen fixation rates were independent of the external oxygen concentration in 3 -3 the range of 0.06 to 0.40 mm mm Equation 1.2 indicates that in order for 0. to remain low, when ^ in 0 is altered, J and or P must also change. In Chapter III the nonex steady-state responses of nitrogen fixation and nodule respiration (J) are examined in response to alterations in O Results from these experiments indicate that immediately after altering the external oxygen concentration both the nitrogen fixation rate and the nodule respiration rate change proportionally with the change in 0^^ as Equation 1.2 predicts. These altered rates, however, are transitory and within a period of several hours both nodule respiration and nitrogen fixation rates return to values similar to those observed under ambient oxygen conditions. It is proposed that a mechanism exists in nodules which regulates the nodule activity in response to altered oxygen concentration in order to both protect nitrogenase and to maintain a constant nitrogen fixation rate. Furthermore, it is concluded that this mechanism does not involve changes in the nodule respiration rate (J), and it is hypothesized that the adaptive mechanism must involve changes in the nodule gas permeability. In Chapter IV, the current methods for measuring nodule gas permeability are discussed. It is found that each of these analytical methods is unsatisfactory and a new procedure is developed. This procedure is then used to measure nodule gas permeability and its

PAGE 12

6 response to altered external oxygen concentrations, and the results of these experiments are reported in Chapter V. It is concluded that soybean nodules regulate the gas permeability in response to the rhizosphere oxygen concentration and thus control the rates of nitrogen fixation. Changing soil temperature was the second condition under which the hypothesis that nodules regulate the gas permeability and are thus capable of controlling nitrogen fixation rates was tested. In Chapter VI, fieldand growth-chamber experiments are reported which indicate that diurnal trends in soybean nitrogen fixation rate are the direct result of daily cycles in soil temperature and are independent of diurnal light cycles. These changes in nitrogen fixation rate v/ere associated with proportional changes in the nodule gas permeability and it is suggested that such changes may reflect an active regulation of the gas permeability by the host plant. The data reported here are consistent with the hypothesis that the nodule gas permeability is under physiological control. Nodule gas permeability was found to be responsive to every environmental stimulus tested which effects nodule activity. Furthermore, these data suggest that the change in nodule gas permeability is not simply a passive response, but one which may be under active regulation. If this is the case nodule gas permeability could play a primary role in both the protection of nitrogenase from oxygen inactivation and in the regulation of nitrogen fixation rates.

PAGE 13

CHAPTER II LONG-TERM EFFECTS OF ALTERED OXYGEN CONCENTRATION ON NITROGEN FIXATION OF INTACT FIELD GROWN SOYBEAN It has been widely reported that symbiotic nitrogen fixation rates are affected by the oxygen concentration in the rhizosphere around plant roots and nodules. Experiments in which nitrogen fixation rates of detached soybean nodules were assayed as either acetylene-reduction or 15 N^ uptake demonstrated that after exposure to altered oxygen concentrations nodule activity responded proportionally to the change in oxygen (Bergersen, 1962; Pankhurst and Sprent, 1975b; Ralston and Imsande, 1982), Similar findings have been reported for attached nodule and root systems of Alnus rubra (Winship and Tjepkema, 1983) and in a wide range of legumes including soybean ( Glycine max), white clover ( Trifolium repens ) pea ( Pi sum sativum ) chickpea (Cicer arietinum), cowpea ( Vigna unguiculata ) peanut ( Arachis hypogoea ) and lupin ( Lupius ablus ) (Ralston and Imsande, 1982; Witty et al 1983; Witty et al., 1984; Winship and Tjepkema, 1985; Weisz et al 1985). Respiration as oxygen uptake by both detached nodules and intact nodulated root systems has also been shown to be sensitive to the oxygen concentration (Tjepkema and Yocum, 1973). Since both nodule respiration and nitrogenase activity are sensitive to the oxygen concentration at which they are assayed, it has been suggested that the oxygen flux crossing a diffusion barrier in the nodule cortex may limit nodule respiration and therefore energy production necessary for nitrogen fixation (Tjepkema, 1971; Tjepkema and Yocum, 1973; Denison et al., 1986) 7

PAGE 14

8 All of the experiments cited above were typically completed in less than an hour after the oxygen concentration around the nodules was altered and thus only the short-term effects of altered oxygen concentration were considered. In contrast to these experiments, there are a number of studies which indicate that when exposed to altered rhizosphere oxygen concentrations for long periods of time, nitrogen fixation rates are unaffected by the change in oxygen concentration. These experiments with long-term exposures to elevated or reduced rhizosphere oxygen concentration have failed to show a response in either 1) field-grown soybean nodule activity as measured by acetylenereduction (Criswell et al., 1976, 1977), 2) plant growth and nitrogen content of soybean and pea (Minchin et al., 1985), or 3) plant and nodule mass in field-grown soybean (Sinclair et al 1985). These data suggest that the role of the rhizosphere oxygen concentration around nodules in limiting nitrogen fixation of intact plants is minimal or non-existent Of the three studies on long-term effects of altered oxygen concentration on nitrogen fixation, Minchin et al. (1985) and Sinclair et al. (1985) inferred fixation rates from growth analysis studies and only Criswell et al. (1976, 1977) assayed nitrogenase activity directly using an acetylene-reduction technique. In the measurements of Criswell et al. (1976, 1977) a nonsaturating concentration of acetylene was used. To calculate the expected ethylene production rate at saturating acetylene concentrations (Vmax) these authors used the Michaelis-Menten equation and an assumed Km. This procedure for predicting Vmax failed to take into account the possible effects of diffusion on the

PAGE 15

9 concentration of acetylene at the reaction site inside the nodule (Winship and Tjepkema, 1983; Denison et al 1983) and therefore may have led to erroneous estimations of nitrogenase activity. The hypothesis that oxygen limits nitrogen fixation rates in symbiotic leguminous systems is supported by numerous short-term oxygen studies and conversely is strongly contradicted by the results inferred from two long-terra growth studies and experiments possibly using an incorrect estimate of nitrogenase activity. In light of this discrepancy the purpose of this research was to evaluate the long-term effects of altered rhizosphere oxygen concentration on nitrogenase activity of intact field-grown soybean plants. Materials and Methods Plant Material And Field Preparation The cultivar 'Biloxi' was field grown in Gainesville Florida, on Arredondo fine sand (loamy, siliceous, hyperthermic Grossarenic Paleudult). Biloxi is an indeterminate, maturity group VIII cultivar which in this experiment remained vegetative until 14 June 1984 when the first flowers appeared. The experiment was terminated on 19 June, at which time the plants were at growth stage Rl (Fehr et al 1971). Field preparation included application of 550 kg ha of 0-10-20 (N-P2O^-K20) fertilizer and the incorporation of 3 L ha of trifluralin (alpha, alpha, alpha, trifluoro-2, 6-dinitro-N, N-dipropyl-p-toluidine ) herbicide. On 2 April, rows spaced 0.9 m apart were seeded with 30 seeds m Open-ended root chambers used for the acetylene-reduction assay (Denison et al., 1983a) were installed in the rows at 1-m

PAGE 16

10 intervals immediately after seeding. Alachlor ( 2-chloro-2 -6 -diethylN-(methyoxYmethyl) acetanilide) and chlorpyrifos (0,0-diethyl-0-(3, 5,6trichloro-2-pyridyl ) phosphorothioate ) were then both applied to the plot at a rate of 4 L ha"''". Sprinkler irrigation was applied to the plot to assure well-watered conditions. Acetylene Reduction Assay In situ ethylene production rates of intact plants in the openended assay chambers were measured at four acetylene concentrations 3 -3 (0.001, 0.004, 0.008, and 0.01 mm mm ) at both mid-day and mid-night over a period of 5 to 7 d. The technique described by Denison et al. (1983b) was used to analyze these data and to determine the mean apparent Km for nitrogenase for all the assays performed (n = 247). The mean and standard error for the Km was found to be 0.00380 and 0.000285 mm"^ mm ^, respectively. This mean value was used to calculate Vmax for each assay as described by Denison and Sinclair (1985). Experimental Design In order to test the long-term effects of altered oxygen concentration around the nodules, two experimental treatments were used. Each treatment consisting of five plants was assayed repeatedly for 2 to 3 d, and then one treatment was switched from ambient oxygen to either 3 -3 0.06, 0.1, 0.3, 0.35, or 0.4 mm mm in mid-afternoon approximately 8 h before the next acetylene-reduction assay. The other five plants were left at ambient oxygen as a control treatment. Plants in both treatment groups were then further assayed twice a day for 2 to 6 more d. In

PAGE 17

11 order to eliminate variability in the Vmax data due to differences among plants in nodule mass, each Vmax estimate was converted to a percentage of the first value observed for a given plant under ambient oxygen conditions. A t-test was used to test for differences between treatment means for normalized Vmax on each day of the experiments. Results On 9 June, five plants which had been continuously assayed for 2 d 3 -3 were switched from ambient to 0.06 mm mm oxygen and assayed for 2 more days. Treatment means for normalized Vmax for this experimental group and a control treatment are presented in Fig. 2.1. After the oxygen concentration was reduced, Vmax appeared to decline; however, this trend started before the change in oxygen concentration occurred and the experimental and control treatment means did not differ significantly (alpha=0.05) on any date before or after 9 June. On 16 June, plants assayed for 3 d at ambient oxygen concentration 3 -3 were switched to 0.1 mm mm oxygen and assayed for another 3 d. Treatment means for normalized Vmax for this experimental group and a control treatment are presented in Fig 2.2. Treatment means for Vmax did not differ at the mid-night assay following the afternoon when the oxygen concentration in the experimental chambers was altered. The following noon treatment means appeared to differ with the plants at 0.1 mm'^ mm oxygen having lower acetylene-reduction rates; however, the apparent difference was not statistically significant ( alpha=0 05 ) Treatment mean values for Vmax did not differ significantly on any date of this experiment.

PAGE 18

12 130r^5120 X o > 110 f 100 c 90 — June J T 8 9 10 11 12 Date Fig. 2.1. Mean normalized maximum acetylene-reduction (percent of initial value) verses date for two treatments, ambient oxygen (open circles), and an ^xygen regulated treatment switched from ambient to 0.06 mm mm oxygen (solid squares) at the time indicated by the arrow. Treatment means did not differ significantly (alpha = 0.05) on any date.

PAGE 19

13 Fig. 2,2. Mean normalized maximum acetylene-reduction (percent of initial value) verses date for two treatments, ambient oxygen (open circles), and an^oxyg^n regulated treatment switched from ambient to 0.1 mm mm oxygen (solid squares) at the time indicated by the arrow. Treatment means did not differ significantly (alpha = 0.05) on any date.

PAGE 20

14 On 17 May, five plants assayed continuously for 2 d were switched 3 -3 from ambient oxygen to 0.3 mm mm Treatment means for normalized Vmax for this treatment group and a control treatment are presented in Fig. 2.3. Acetylene reduction rates did not differ between treatments 3 -3 over the 24-h period these plants were assayed at 0.3 mm mm oxygen. On 18 May, the chamber oxygen concentration was further increased to 3 -3 0.35 mm mm and the plants assayed for another 24 h. On the night of 19 May the experimental treatment mean for Vmax was significantly lower (alpha=0.05) than that of the control plants. Treatment means did not differ significantly on any other date. On 28 May, five plants assayed for 2 d were switched from ambient 3 -3 to 0.4 mm mm oxygen and assayed for 6 more days. Data from these plants are presented in Fig. 2.4. Treatment means for normalized acetylene-reduction did not differ between oxygen treatments except on 3 -3 one occasion 6 d after the 0.4 mm mm oxygen treatment began. Discussion In these experiments acetylene-reduction rates for intact fieldgrown plants were assayed at approximately mid-day and mid-night over a period of approximately 1 wk. At about m.id-week, the oxygen concentration in the rhizosphere around five plants was altered about 8 h before the mid-night assay. Thus a response to the altered oxygen concentration which might have occurred in the first 8 h of the treatment would not have been observed. There were no long-term (greater than 8 h) effects of lowering the rhizosphere oxygen concentration to either 0.06 or 0.1 mm'^ mm~^ on acetylenereduction rate.

PAGE 21

15 Fig. 2.3. Mean normalized maximum acetylene-reduction (percent of initial value) verses date for two treatments, ambient oxygen (open circles), and an oxygen regulated treatment switched from ambient to 0.30 mm^ mm ^ and 0.35 mm"^ mm ^ oxygen (solid squares) at the times indicated by the arrows. An "*" indicates treatment means differ significantly at the alpha = 0.05 level.

PAGE 22

16 Fig. 2.4. Mean normalized maximum acetylene-reduction (percent of initial value) verses date for two treatments, ambient oxygen (open circles), and an^oxyg^n regulated treatment switched from ambient to 0.4 mm mm oxygen (solid squares) at the time indicated by the arrow. An "*" indicates treatment means differ significantly at the alpha = 0.05 level.

PAGE 23

17 Similarly, there vi/ere no consistent long-term effects of elevated oxygen 3 -3 concentration. On one date, plants at 0.3 mm mm oxygen had significantly lower acetylene-reduction rates but this reduction was not consistent across time. That elevated oxygen concentration did not result in long-term effects on nitrogenase activity is further demonstrated in Fig. 2.4, where plants switched to a rhizosphere oxygen concentration of 0.4 mm"^ mm did not differ from the controls in nitrogenase activity over the subsequent 5 d. These data confirm the findings of Criswell et al (1976, 1977), Minchin et al. (1985) and Sinclair et al. (1985), and further demonstrate that at least after an 8-h of exposure to altered oxygen concentration in the range of 0.06 to 0.4 mm"^ mm ^ there are no long-term effects of oxygen concentration on nodule activity. These data and the data from other long-term oxygen studies indicate that nitrogen fixation rate in symbiotic leguminous systems is not oxygen limited. Such a conclusion is in contradiction to the numerous short-term studies of nodule activity in many legumes exposed to altered oxygen concentration. The shear number of such short-term experiments and the diversity of experimental procedures used in them, make it unlikely that each would contain some common confounding and unknown factor. Criswell et al (1976) suggested that a mechanism may exist in soybean nodules which allows them to adapt to a wide range of soil oxygen concentration. If such an unknown mechanism exists in nodules and if it took several hours to respond to altered oxygen conditions this could explain the discrepancy between the longand short-term responses to altered oxygen concentration.

PAGE 24

CHAPTER III NONSTEADYSTATE NODULE RESPIRATION AND NITROGEN FIXATION IN RESPONSE TO ALTERED OXYGEN CONCENTRATION. The long-term oxygen studies reported in Chapter II indicated that when the rhizosphere oxygen concentration around intact field grown soybean plants was altered over a range of 0.06 to 0.4 mm"^ nun ^, nitrogenase activity was independent of the external oxygen concentration at least after an initial 8-h period. A serious implication of these data on nitrogenase activity at elevated oxygen concentrations concerns nitrogenase protection from oxygen inactivation Under ambient oxygen the internal oxygen concentration in the infected cells of soybean nodules has been calculated to be on the order of -7 3-3 2*10 mm mm (Appleby, 1984). This low internal oxygen concentration is consistent with the fact that nitrogenase is inactivated by oxygen (Robson and Postgate, 1980). Tjepkema and Yocom (1974) demonstrated that the major source of resistance to oxygen diffusion into soybean nodules and the site of the decrease in the oxygen concentration from ambient levels to this extremely low internal partial pressure is a diffusion barrier in the inner cortex of the nodule. Assuiraning the major resistance to inward diffusion of oxygen is in the nodule cortex, Sheehy et al. (1983) modeled the flux of oxygen into nodules using Pick's first law as = ^ ^ex Oin) (3-1) 18

PAGE 25

19 3 -2 ~1 where J is the flux (mm mm s ) of oxygen crossing the diffusion barrier (assumed to be proportional to the nodule respiration rate), P is the gas permeability of the diffusion barrier (mm s ''"), O^^ is the rhizosphere oxygen concentration (mm"^ mm ) and O^^ is the oxygen 3 3 concentration (mm mm ) in the intercellular air spaces around the bacteroid infected cells inside of the diffusion barrier. As 0^^ is many orders of magnitude below 0^^ Equation 3.1 can be simplified. J = P O (3.2) ex In the experiments described in Chapter II, 0 was increased from 3-3 3-3 ambient (0.2 mm mm ) to as high as 0.4 mm mm oxygen without any long-term effects on nodule activity. Equation 3.2 predicts that for a change in 0 a proportional change in J will also occur unless the permeability of the nodule (P) changes as well. Thus, in order to maintain a low internal nodule oxygen concentration for nitrogenase protection when the rhizosphere oxygen concentration (0^^^) is altered, either the nodule respiration rate (J) or the nodule gas permeability (P) must also change. Bergersen and Turner (1975, 1980) reported that isolated bacteroids from soybean nodules may contain from two to four different cytochrome oxidase systems which can function at different internal nodule oxygen concentrations. These authors demonstrated that bacteroid respiration rate can vary over a range of dissolved oxygen concentrations from approximately 3*10 mm^ mm ^ to 3*10 mo? mm and yet maintain a fairly constant supply of ATP to nitrogenase for nitrogen fixation.

PAGE 26

20 This implies that nodule respiration might vary in response to altered oxygen concentrations and yet nitrogenase activity could remain constant. Such a mechanism is consistent with Equation 3.2 and could explain the results reported for long-term exposures to altered rhizosphere oxygen concentration. A stable low internal nodule oxygen environment would be maintained by an altered respiration rate which would continue to supply nitrogenase with the same amount of ATP resulting in constant acetylene-reduction rates. The purpose of this research was to test the hypothesis that when exposed to long-term alterations in rhizosphere oxygen concentration soybean nodules maintain a low and stable internal oxygen environment by regulating their nodule respiration rates. Also, it is further hypothesized that nitrogen fixation rates as m.easured by acetylene-reduction are unaffected by changes in oxygen concentration. This predicts that when the external oxygen concentration around soybean nodules is either increased or decreased a proportional change in the oxygen uptake rate will occur without a concomitant change in acetylene-reduction rate. Materials and Methods Hydroponically Grown Plant Material To be consistent with the field experiments reported in Chapter II the cultivar 'Biloxi' was used for these experiments. Seeds were surface sterilized with 2% sodium hypocholorite and germinated on moist filter paper. After germination the seedlings were transferred to growth pouches until the primary root was approximately 50 mm long. The seedlings were then individually transferred to bored #3 rubber stoppers

PAGE 27

21 which were placed in the lid of a 1.5 L hydroponic chamber made from 102 mm diameter PVC pipe (Fig. 3.1) and inoculated with a commercial Brady rhizobium japonicum inoculum (Nitragin Corporation). Plants were maintained in half -strength nitrogen-free nutrient solution (Imsande and Ralston, 1981) which was continuously aerated by passing 33 mL s of air through an aquarium glass bead bubbler in the bottom of each growth chamber. The growth chambers were submerged in a water bath which regulated the temperature around the roots and nodules to 26 C. Illumination was provided by a "Sun-Brella" (Environmental Growth Chambers, Chagrin Falls, Ohio 44022) which consisted of a multi-vapor metal halide lamp (General Electric #E-37) in combination with one highpressure sodium lamp (General Electric #E-18) in a water-cooled jacket 2 -1 and which provided 950 to 1400 uE/m s of photosynthetically active radiation depending on position in the canopy. The photoperiod was adjusted to a 16-h day to assure that plants remained vegetative throughout the assay period. The day before an individual plant was to be assayed it was transferred from the hydroponic growth chamber to a stainless-steel flow-through assay chamber (Fig. 3.2). An intact plant, including the rubber stopper through which the stem grew in the hydroponic chamber, was used by inserting the nodulated root system through the bore of the assay chamber until the rubber stopper sealed the top of the chairiber (see Fig. 3.2). An air-tight seal at the lower opening of the chamber was made by impregnating the roots in silicon grease (Dow Corning) inside of a split and bored #3 rubber stopper. The assay chamber was then placed on top of a modified hydroponic chamber. Moisturized air

PAGE 28

Fig. 3.1. Hydroponic growth chamber made from 102 mm diameter PVC pipe and end cap.

PAGE 29

Fig. 3.2. Stainless steel flow through assay chamber (A) with hydroponic support chamber (B). Moisturized gas was supplied to the assay chamber through port D exited the assay chamber from port C, Air entered the hydroponic chamber through port E. The lower opening of the assay chamber was sealed with a split and bored #3 rubber stopper (G). Roots passing through the stopper were impregnated in silicon grease (F), and the stopi.)er was further capped with putty (H).

PAGE 30

24 was continuously passed through the assay chamber at a rate of 0.33 inL s ^ to assure that the nodules were well aerated. The assay chamber was maintained in the same water bath and under the same lighting conditions as the growth chambers described above. Nodule Respiration; Oxygen Uptake The oxygen concentration in the gas phase was measured with a Walker-type Clark-style oxygen electrode (Delieu and Walker, 1981). Root plus nodule respiration was estimated as the difference between the oxygen concentration in the air supplied to the chamber and that in the gas exiting the chamber. When an experiment was com.pleted, the plant in the assay chamber was removed and the nodules harvested from the root system. The plant was then returned to the assay chamber, and the bare root respiration rate was assayed under the experimental conditions used for that plant. Bare root respiration rate was then subtracted from the root plus nodule respiration rates to yield nodule respiration. Nitrogenase Activity Nitrogenase activity of the intact nodules in the assay chamber was determined using the acetylene-reduction technique. To do this the air supply rate to the assay chamber was increased to 3.33 mL s and 3 —3 acetylene added to a final concentration of 0.10 mm miti To determine the nodule ethylene production rate, 4 min after the acetylene flow was turned on a 1 mL sample was drawn from the gas exiting the assay chamber and injected into a gas chromatograph fitted with a flame ionization detector. The acetylene was then removed from the air supply which was

PAGE 31

25 returned to the normal flow rate and any residual acetylene flushed from the chamber Saturating concentrations of acetylene have an inhibitory effect on nitrogenase activity in which the ethylene production rate begins to decline after a period of several minutes of exposure (Minchin et al 1983). This inhibitory effect can cause a serious under estimation of nitrogenase activity depending on the amount of time which has elapsed between when the nodules were initially exposed to the high concentrations of acetylene, and when the ethylene production rate was assayed. Therefore, it was necessary to determine the time dependence of this acetylene response for the hydroponically grown soybeans used in this experiment. To do this acetylene was added to the gas supplied to 3 -3 the assay chamber to a final concentration of 0.10 mm mm and the gas exiting the chamber was sampled every two minutes starting 1 min after the addition of acetylene for a total of 15 min. Nine assays were run in this fashion. Non-Steady-State Experimental Design The night before a plant was to be assayed it was transferred from a hydroponic growth chamber to the assay chamber. The following morning nodule activity and respiration were assayed at ambient oxygen. The oxygen concentration in the gas supply was then altered to either 0.1 3 -3 (for 10 plants) or 0.4 mm mm (for four plants). The time requirec for the step change in oxygen concentration to be completed in these experiments was approximately 5 min. Nodule respiration and acetylenereduction were then repeatedly assayed until steady-state was reached.

PAGE 32

26 VvTien the experiment was terminated the nodules were harvested and the root respiration assayed at each oxygen concentration used for the individual experiment. Results Time Course for Acetylene Inhibition of Nitrogenase Activity Nitrogenase activity as a percent of the maximxam rate of ethylene production observed is plotted against time since the addition of acetylene to the assay chamber for nine individual assays in Fig. 3.3. 3 -3 The concentration of acetylene in the chamber reached 0.10 mm mm after 3 min and maximum ethylene production occurred about 1 min later or a total of 4 min after the addition of acetylene to the gas supply. Acetylene-reduction rates were then stable for approximately 5 more minutes and then began a slow decline until they were 95% of the maximum rate at 14 min. 3 -3 Respiration and Acetylene-Reduction in Response to 0.1 mm mm Oxygen A total of 10 plants were assayed at ambient oxygen and then 3 -3 sv/itched to 0.1 mm mm oxygen and continually assayed until steadystate was attained. Nodule respiration and maximum acetylene-reduction rate as a percent of that observed under ambient conditions are plotted against time since the growth chamber lights came on in the morning for four representative plants in Fig. 3.4. Nodule respiration rate decreased dramatically when the oxygen concentration was dropped to 0.1 mm"^ mm and then recovered slowly over the next few hours. Acetylene reduction rate also decreased when the oxygen concentration was lowered

PAGE 33

27 Time (minutes) Fig. 3.3. Ethylene production rate as a percent of the maximum rate achieved verses time since the chamber was switched from CO 3 -3 to 0.1 mm mm acetylene.

PAGE 34

28 Time (hours) Fig. 3,4. Nodule respiration (dots) and maximum acetylene-reduction (open squares) as a percent of the initial value verses time since the growth chamber lights came on for individual plants assayed under ambient conditions and then switched to 0.1 mm mm oxygen Arrows indicate the time when the oxygen concentration was altered.

PAGE 35

29 but this decrease was also transitory as there was a recovery to rates similar to those observed under ambient oxygen. A summary of the final steady-state values for nodule respiration and acetylene-reduction rates 3 -3 at 0,1 mm mm oxygen is presented in Table 3.1. Table 3.1. Steady-state values at 0.1 mm mm oxygen as percent initial value at ambient oxygen concentration. Expt, % Acetylene % Nodule # Reduction Respiration 1 98 121 2 91 80 3 92 104 4 91 5 89 106 6 92 112 7 104 111 8 94 109 9 90 101 10 99 107 MeantS.E: 94.3+1.7 104.23,6 Respiration and Acetylene Reduction Rates in Response to 0.4 mm mm Oxygen Nodule respiration and acetylene-reduction were assayed for four 3 -3 plants under ambient oxygen and then switched to 0.4 mm mm oxygen. These data are presented in Fig. 3.5. Nodule respiration decreased after increasing the oxygen concentration around the nodules and then recovered to rates comparable to those initially observed. Non-steadystate acetylene-reduction rates were measured for two of these plants and like nodule respiration initially decreased after increasing the oxygen concentration. Final steady-state acetylene-reduction rates for

PAGE 36

30 120 I) — — I — f t > I I — — rc o O-i g 80 if) Dark ^( — 1 r I I III 1 I — r120 80 Dark o 40 5" -4 1 f t t 1 • 1 1 > I ^ll • ' I I I I 1 t 1 1 < I I I I > 1 120 o < 80 3 Q X 40 10 20 30 10 20 30 Time (hours) Fig. 3.5. Nodule respiration (dots) and maximum acetylene-reduction (open squares) as a percent of the initial value verses time since the growth chamber lights came on for individual plants assayed^under ambient conditions and then switched to 0.4 mm mm oxygen. Arrov/s indicate the time when the oxygen concentration was altered.

PAGE 37

31 all four plants were similar to rates initially observed at ambient oxygen. A summary of these steady-state values for nodule respiration 3 -3 and acetylene-reduction at 0.4 mm mm oxygen is presented in Table 3.2. Table 3.2. Steady-state values at 0.4 mm mm oxygen as percent initial value at ambient oxygen concentration. Expt. # % Acetylene Reduction % Nodule Respiration 1 89 96 2 92 99 3 102 98 4 100 90 MeantS .£: 98.83.9 95.02.0 Discussion Acetylene reduction rates for these intact hydroponically grown soybean plants were assayed four minutes after acetylene was added to the gas supply at which time any residual acetylene was flushed from the assay chamber. This was approximately 5 min before any inhibitory effects of acetylene on nodule activity could be observed (see Fig. 3.3). Minchin et al. (1983) demonstrated that saturating concentrations of acetylene can result in a reduction in nitrogenase activity. The data presented here for soybean, however, indicate that this inhibition was very small being 5% or less over a 14-min exposure to 0.10 mm mm acetylene. For this reason it was concluded that when assayed in the fashion used in these experiments the acetylene-reduction assay did not have an inhibitory effect on nodule activity.

PAGE 38

32 As in the field exxjeriments reported in Chapter II, these hydroponic plants exhibited steady-state acetylene-reduction rates which were independent of the external oxygen concentration from a range of 0.1 to 0.4 mm^ mm ^. There were, however, short-term responses of nitrogenase activity to altered oxygen concentrations. Decreasing the 3 -3 oxygen concentration to 0.1 mm mm resulted in decreased ethylene production rates and a dramatic decrease in nodule respiration. These decreased rates were transitory and full recovery of both acetylenereduction and nodule respiration rates were complete in 4 to 6 h. These data indicate that given enough time nodules can adapt to decreased rhizosphere oxygen concentrations. The fact that steady-state nodule 3 -3 respiration rates at 0.1 mm mm oxygen were similar to those observed under ambient conditions does not support the hypothesis that the mechanism for maintaining a constant nitrogen fixation rate under varying oxygen concentrations involves an alteration in nodule respiration rate. 3 -3 Exposure to 0.4 mm mm oxygen resulted in short-term reductions in both nodule respiration and acetylene-reduction rates. Such a response to increased oxygen concentrations may have been caused by partial oxygen inactivation of nitrogenase. The data from the plants assayed at 0.1 mm"^ mm oxygen indicate that the nodule response time to altered oxygen concentrations was several hours. In these experiments the assay chamber oxygen concentration was changed in 5 min. Such a rapid increase in the external oxygen concentration may have overwhelmed the adaptive system. During the period that the oxygen concentration in the chambers was left at these supra-ambient levels, the nodule

PAGE 39

33 respiration and acetylene-reduction rates recovered and the final steady-state rates were comparable to those initially observed. Thus, while the external oxygen concentration had been doubled, the nodule respiration rates remained the same. These data are also inconsistent with the hypothesis that the internal low nodule oxygen concentration was maintained by an alteration in nodule respiration rate. These data confirm that intact soybean nodules have a mechanism for adapting to a wide range of altered rhizosphere oxygen concentrations. This adaption mechanism is time dependent and requires several hours to complete adjustment to the new external oxygen environment. Contrary to the proposed hypothesis, this adaptivity does not involve a change in nodule respiration rate; in fact nodule respiration seems to be as equally affected by this mechanism as does nitrogenase activity. J

PAGE 40

CHAPTER IV QUANTITATIVE APPROACHES TO MEASURING NODULE GAS PERMEABILITY: DEVELOPMENT OF A RAPID NON-DESTRUCTIVE ASSAY The data presented in the last two chapters indicate that soybean nodules contain a mechanism for maintaining a stable low internal oxygen concentration when the nodules are exposed to altered rhizosphere oxygen concentrations. It was hypothesized in Chapter III that this mechanism would involve the regulation of either the nodule respiration rate or the nodule gas permeability. Data presented did not indicate that alteration of nodule respiration was involved in the adaptive mechanism, as nodule respiration and nitrogen fixation rates appeared to respond in similar fashions to altered oxygen concentrations. Since it is likely that the majority of nodule respiration and all of nitrogen fixation take place inside the diffusion barrier, changes in the oxygen flux across this barrier either due to an altered oxygen gradient or to a change in the gas permeability of the barrier could affect both equally. It is hypothesized that the mechanism involved in nodule adaption to altered oxygen concentrations consists of regulation of the gas permeability of the diffusion barrier in the nodule cortex. In order to test this hypothesis a reliable method for measuring the nodule gas permeability is essential. Two of the methods currently used for measuring nodule permeability are reviewed here, and it is concluded that each is unsatisfactory. A new method is developed which can be used with intact plants and which results in estimates of both nitrogenase activity and the nodule gas permeability. This technique is 34

PAGE 41

35 then used with intact hydroponically grown soybean plants and the resultant estimated values of nodule gas permeability compared v/ith estimates from other sources. Review of Current Techniques Nodule Respiration and Fick's First Law Approximation Sheehy et al. (1983) modeled whole nodule respiration using a steady-state one-dimensional diffusion equation similar to Equation 3,1 but rearranged to solve for nodule gas permeability such that P = J / (O O. ) (4.1) ex in where P is the gas permeability (irm s ) of the diffusion barrier, J is 3 -2 -1 the oxygen flux density crossing the nodule cortex (mm mm s ), O^^ 3 -3 is the external oxygen concentration (mm mm ) and O^^ is the oxygen concentration in the intercellular air spaces internal to the cortical 3 -3 barrier (mm mm ) Since 0^^ is experimentally defined, P can be estimated from Equation 4.1 if the oxygen flux (J) into the nodule interior is known and given certain assumptions about O. Since 0. i in in many orders of magnitude lower than 0^^ its absolute value is not very 3 -3 significant and these authors assiamed^it to be equal to 0.001 mm nim oxygen To estimate J in Equation 4.1, Sheehy et al (1983) measured nodul carbon dioxide evolution and made certain assumptions to convert this into an oxygen flux. The first assumption concerned the fraction of total respiration representing a respiratory flux across the diffusion

PAGE 42

36 barrier. Not all the carbon dioxide evolved by nodules is from the nodule interior, as cells of the inner and outer cortex are also respiring. Quantitative measurements of the respiration rate of the cortical layers have not been made, although Sprent (1972) reported that cells in the nodule cortex show intense metabolic activity. The vascular tissue is also located in the inner cortex, so respiration associated with the loading, unloading and transport of carbon and nitrogen compounds must take place in this region. Witty et al. (1983) estimated the percent of nodule respiration that remained when nitrogenase activity was extrapolated to zero to be about 20% of the total carbon dioxide evolved under ambient conditions. Sheehy et al. (1983) then used this figure to represent the fraction of total respiration taking place in the nodule cortex, or outside the diffusion barrier and thus set J in Equation 4.1 to be equal the measured respiration rate minus this value. This assumes 1) that little or no respiration associated v;ith nitrogen assimilation takes place in the nodule cortex, 2) that respiration associated with cortical or vascular transport is minimal, and 3) that respiration associated v/ith the nodule interior other than that associated with nitrogen fixation is negligible. Errors in any of these assumptions will lead to proportional errors in the resultant value of P from Equation 4.1. Furthermore, should the ratio of cortical to interior respiration vary across experimental treatments, this method of determining P will yield an erroneous change in P. To estimate J from carbon dioxide evolution rates it is also necessary to know the relationship between oxygen uptake and carbon

PAGE 43

37 dioxide evolution. Sheehy et al. (1983) assumed that the respiratory quotient (ratio of carbon dioxide evolved to oxygen uptake) was constant across time and experimental treatments, and equal to unity. Bergersen (1971) reported that the respiratory quotient for soybean nodules decreased from 1.31 to 1.05 as the external oxygen concentration around 3 -3 the nodules was increased from 0.008 to 0.24 mm mm Winship and Tjepkema (1985) also reported that the respiratory quotient for Alnus rubra nodules decreased form 1.2 to 1.0 as the external oxygen 3 -3 concentration was elevated from 0.2 to 0.4 mm mm These reported respiratory quotients are highly variable and are apparently sensitive to at least one of the experimental treatments under which nodule gas permeability has been estimated with this method. For example, Sheehy et al. (1983) used this estimation procedure and reported that P appeared to be responsive to changes in the external oxygen concentration around white clover nodules. Vvhen O was increased from ex 3 -3 0.21 to 0.8 mm mm the estimated nodule permeability decreased by 76%. If it is assumed however, that the respiratory quotient was decreasing (as Bergersen, and Winship and Tjepkema 's data indicate) then as the oxygen concentration around these white clover nodules was raised, the estimated P's would indeed appear to be lower at higher values of 0^^. The assumption of a constant respiratory quotient is apparently a serious limitation to the use of this model for estimating nodule gas permeability. It was concluded that this analysis procedure is unsatisfactory for testing the hypothesis that nodule gas permeability is involved in the adaptive mechanism to altered rhizosphere oxygen concentrations

PAGE 44

38 Combined Model Using the Diffusion and Michaelis-Menten Equations Winship and Tjepkema (1983) and Denison et al (1983b) developed a technique for estimating nodule permeability based on the acetylenereduction assay. The model assumes the reduction of acetylene to ethylene by nitrogenase can be modeled using the Michaelis-Menten equation if the effect of the external diffusion barrier on the concentration of acetylene at the enzyme site is known. A detailed development and analysis of the derivation of this model, has been presented by Johnson and Thornley (1985). While each of these authors used differing algebraic forms of the model one representative equation which is equivalent to those presented in all three of the above citations is V = (Vmax + KM*k + k*A ) / 2 ex ( (Vmax + KM*k + k*A ) 4*k*Vmax*A ) / 2 (4,2) ex ex where V is steady-state nodule ethylene production rate (mm"^ s ^), Vmax is the maximum ethylene production rate attained at saturating 3 -1 concentrations of acetylene (mm s ) KM is the apparent nitrogenase Michaelis-Menten constant for acetylene divided by the solubility of 3 ~ 3 3 ~ 1 acetylene (mm mm ) k is the nodule gas conductivity (mm. s the product of nodule gas permeability (P) times nodule surface area), and 3 -3 A is the external concentration of acetylene (mm mm ). If the value of V is known at a series of different A 's then the three unknown ex parameters, Vmax, KM, and k can be solved for using a non-linear regression (Winship and Tjepkema, 1983; Denison et al. 1983b; Denison

PAGE 45

39 and Sinclair, 1985). The nodule gas permeability can then be calculated once the total nodule surface area has been estimated such that P = k / Narea (4.3) 2 where Narea is the total nodule surface area (mm ) The nodule surface area can be estimated by the method of Weisz et al. (1985) using measurements of the cross-sectional dimensions of the nodules. Use of the above approach to determine P is, however, dependent on the use of the acetylene-reduction assay which as described in Chapter III may inhibit nitrogenase activity and thus result in substantial under estimations of Vmax (Minchin et al 1983; Witty et al 1984; Minchin et al 1985). To avoid these acetylene effects, Denison et al. 3 -3 (1983b) used a series of acetylene concentrations below 0.01 mm mm which is an order of magnitude below those at which Minchin et al. (1983) reported the acetylene effects. While this approach avoids the confounding effects of acetylene on the parameters being assayed, it limits the range of acetylene concentrations over which V can be assayed. Thus the regression procedure to solve for the three unknown parameters in Equation 4.2 uses data obtained from only a very small portion of the total acetylene response curve. Furthermore, if only four concentrations of acetylene are used as done by Weisz et al (1985) only one degree of freedom for the error term in a nonlinear regression is left and the ability of the procedure to estimate the parameters with a significant degree of reliability is severely restricted. To improve upon this situation, Denison and Sinclair (1985) assumed that the K14 was

PAGE 46

40 3 -3 constant and equal to 0.0035 iran mm This resulted in a reduced form of the model such that V = (Vmax + 0.0035*k + k*A ) / 2 ex ( (Vmax + 0,0035*k + k*A y 4*k*Vmax*A ) / 2 (4.4) ex ex This provided one more degree of freedom for the error term in the regression procedure. If, however, the KM is not constant or the wrong KM value is used in Equation 4.4, eliminating it from the full model may introduce systematic error into the predicted values of Vmax and k. Sensitivity of the Reduced Model to Errors in the Assumed KM To test whether Equation 4.4 would predict correct values of k and Vmax if the true KM were indeed variable, a series of nine data sets were simulated. Equation 4,2 was used to generate values of V at four concentrations of acetylene (0.001, 0.004, 0.008, 0.01 mm"^ mm ) using three values of KM (0.002, 0.004, and 0.008 mm'^ mm ) three values of k 3-1 3-1 (6.0 12.0 and 24.0 mm s ), and one value of Vmax (0.15 mm s ). These values were similar to those reported for intact soybean nodules (Weisz et al., 1985). Then Equation 4.4 was used with a nonlinear regression to solve each of the simulated data sets for a predicted k and Vmax assuming a KM value of 0.004 mm"^ mm ^ The results of this simulation are presented in Table 4.1.

PAGE 47

41 Table 4.1. Sensitivity of the reduced model to the assumed KM. CORRECT 1 TTlTn TTITTI I V ilULl IILIII / CORR|;CT_j^ ( TTITTI Q ^ % ERROR IN THE k PREDICTED: Vmax 0.002 0.002 0 002 6.0 12.0 24 0 4.7 13.3 42 .9 30.2 22.3 16.7 0.004 0.004 0.004 6.0 12.0 24.0 0.0 0.0 0.0 0.0 0.0 0.0 0.008 0.008 0.008 6.0 12.0 24.0 -10.5 -34.3 -36.3 -23.7 -3.7 -18.5 Percent error in k and Vmax estima^es_^sing an assumed KM of 0.004 mm"^ mm and fixed Vmax of 0.15 mm s When the reduced model was used to predict k and Vmax from these simulated data sets the values of the predicted parameters were highly accurate when the correct KM was used. When the KM was not constant but 3 -3 varied from the assumed value of 0.004 mm mm the predicted nodule conductances, and maximum rates of ethylene production deviated by -36.3 to 42.9% from the correct values. Large errors can result from the use of this model to predict k and Vmax if the assumed KM is in error, or varies over the experimental treatments being used. This represents a serious limdtation to the use of Equation 4.4 to predict nodule conductance Reduced Model with Fixed Vmax In contrast to the approach of Denison and Sinclair (1985) who fixed the value of KM in Equation 4.4 and then solved for k and Vmax, Winship and Tjepkema (1983) measured ethylene production rates of Alnus

PAGE 48

42 rubra nodules at a series of acetylene concentrations including saturating levels. The observed value attained at saturating acetylene was assumed to represent Vmax which was substituted into Equation 4.2. A non-linear regression was then used to solve for KM and k. In this manner all three parameters were allowed to vary from plant to plant and across treatments. The method of Winship and Tjepkema (1983) for determining KM, k and Vmax was tested for soybeans using intact hydroponically grown plants. The plant growth conditions and assay chamber were identical to those described in Chapter III. Acetylene-reduction was assayed at 0.10 3 -3 mm mm acetylene as described in Chapter III, except instead of flushing the chamber with fresh air after exposure to saturating acetylene, the acetylene concentration was altered in order to additionally measure ethylene production at 0.001, 0.005, 0.015 and 3 -3 0.025 mm mm acetylene. The observed ethylene production rate at 0.10 3 -3 mm mm acetylene was assumed to represent Vmax and was substituted into Equation 4.2 which was used with the remaining data to solve for k and KM. A total of 15 plants were assayed in this fashion at ambient 3 -3 oxygen, ten were assayed at 0.1 mm mm oxygen and two v/ere assayed at n / 3 -3 0.4 mm mm oxygen. Observed ethylene production rates for a representative plant assayed at ambient oxygen at four acetylene concentrations are shown in Fig. 4.1. The "best fit" results from the non-linear regression for KM and k were used to generate the lower line which results in an R of better than 0.999. Surprisingly, choosing a value of k which was 50% higher than the one obtained by the non-linear regression could also be used to generate a curve which approximated the actual data if the KM

PAGE 49

43 C2H2(nnm^mm3)x102 Fig. 4.1. Ethylene production verses acetylene concentration for an intact hydroponical ly grown soybean (dots). Lines represent curves generated by Equation (4.2) using the correct Vmax, and two different KM's and values of k which differ by 50%. Error bars represent 95% confidence intervals for the individual ethylene production rates and acetylene concentrations

PAGE 50

44 was also increased. This second curve is the upper line in the figure 2 and the resultant R for this fit is also better than 0.999. Both of these curves are within the 95% confidence intervals for the individual data points. A family of curves each with different values of k and KM can be fit to the data using Equation 4.2, each of which will closely approximate the data and result in a low error sum of squares. The curve which the non-linear regression predicts as a best fit may, therefore, be more a factor of random or systematic error in the data than an actual indication of the true value for k. This represents a serious limitation to the use of Equation 4.2 for prediction of nodule conductance even if Vmax is known. Validation of Vmax Estimates for the 1984 Field Data In Chapter II and VI estimates of Vmax from field studies using the reduced model with an assumed KM are reported. It has been demonstrated that this model can not be used to predict accurately the nodule conductance, k, if the KM is unknown or changing over the experimental treatments. Also the simulations reported in Table 4.1 indicate that depending on the degree of error in the assumed KM and the true value of the nodule conductance, the resultant estimation of Vmax from the reduced model might also be incorrect. To test the validity of the Vmax estimates from the field studies the data from the 27 assays of intact hydroponically grown plants 3 -3 assayed at 0.1, 0.2 and 0.4 mm mm oxygen described above were used. The analysis procedure used with the field plants (see Chapter II) was repeated with these data such that Equation 4.2 was used to estimate a

PAGE 51

45 mean KI-' Ici. the ertire data set. This mean KM was then used in Equation 4.4 to estimate the k and Vmax associated with each assay using the ethylene production rates observed at the three lowest acetylene 3 -3 concentrations which were 0.001, 0.005 and 0.015 mm mm In this way the estimated Vmax could be compared with the observed maximum rate measured at saturating acetylene concentrations. The results from these assays are reported in Fig. 4.2 where estimated Vmax is plotted against 2 the observed maximum rate. There is a good linear fit (R =0.91) and the slope of the regression is 0.80 which indicates that the estimated values of Vmax are about 20% under estimated. While the absolute value of the estimates is low, the error is consistent across the full range of rates measured. In light of this, Vmax estimates reported in Chapters II and VI were presented as percent of an initial value observed under control conditions. This de-emphasized the absolute value of the estimates and yet allowed comparisons of Vmax across treatments Conclusions In theory Equation 4.2 can be used to predict the Vmax, k, and KM for nitrogen fixing nodules. However, as acetylene may inhibit nitrogenase activity, unless the assay can be completed rapidly it is necessary to use concentrations well below those at which nitrogenase is saturated. This results in data sets which are difficult to resolve into reliable estimates with a non-linear regression. If the KM is assumed to be constant and known, then the model can be used to predict Vmax and k even at these low acetylene concentrations. If however

PAGE 52

46 6i 1 1 r Vmax (mm3/s)xi Fig. 4.2. Estimated values of Vmax verses measured maximum rates of acetylene-reduction ^or plants assayed under ambient and either 0.1 or 0.4 mm mm oxygen. Estimates were calculated from eth^lene^production rates measured at 0.001, 0.005 and 0.015 mm mm acetylene.

PAGE 53

47 the assumed KM is in error or changes during the experiment, large errors in k and Vmax may occur. Finally, the shape of the acetylene response curve for various values of k and KM at a given Vmax reflects an interaction between these parameters, and various values of k and KM can result in similar curves over the concentration range used. Unless the KM is known for each experimental treatment, this model may give erroneous values for k. It is concluded that this model can be used with caution to predict Vmax, but is unreliable for predicting the nodule gas conductance and is therefore unsatisfactory for testing whether the nodule gas permeability adapts to altered rhizosphere oxygen concentrations Development of the Non-Steady-State Lag Phase Model Davis (1984) suggested that the time required for nodules to reach steady-state ethylene production rates after being exposed to acetylene should be a function of both the geometry of the nodules and the diffusivity of the nodule tissue. Davis (1984) measured this "lagphase" for acetylene-reduction in detached nodules from Sesbania rostrata and hairy winter vetch and calculated the time required to reach steady-state ethylene production. To compare reaction times from nodules of different shapes or sizes, Davis (1984) assumed that nodules were homogeneous and nonreactive through out. Both of these assujtiptions he admitted are incorrect, as the nodule cortex represents the majority of the resistance to inward gas diffusion and the nodule interior reacts with acetylene. Furthermore, a method for relating the time required to achieve steady-state to nodule gas permeability was not presented.

PAGE 54

48 Davis (1984) demonstrated that the lag time to steady-state ethylene production rate does vary from species to species and is dependent on the pretreatm.ent of the nodules used, indicating that the lag time may be a valid indicator of the nodule gas permeability. A method is presented here which can be used to 1 ) measure the lag time to steadystate ethylene production in intact nodules, 2) calculate the time constant for this lag period, and 3) estimate the mean nodule gas permeability of the nodules assayed from this time constant. Physical Description of the Model System The diffusion characteristics and anatomy of soybean nodules have been previously described in detail (Spent, 1972; Bergersen and Goodchild, 1973; Tjepkema and Yocum, 1974; Goodchild, 1977; Sinclair and Goudriaan, 1981; Selker and Newcomb, 1985; Sinclair et al 1985). In this model nodules are assumed to be spherical with the majority of the resistance to gaseous diffusion occurring in the nodule cortex which corresponds to an outer layer of the sphere. Internal to this diffusion barrier a system of intercellular air spaces permeates the nodule and therefore, the diffusivity of gases in this region is rapid compared to the diffusion rate through the cytoplasm of the cortical zone. In light of this high rate of internal diffusivity, Sinclair and Goudriaan (1981) predicted that the concentration of gases in the nodule interior would be nearly uniform. It is assumed therefore that the concentration of acetylene and ethylene in the nodule interior is in equilibrium with the inner surface of the diffusion barrier. Once inside the barrier, acetylene reduction to ethylene can be described by Michaelis-Menten kinetics and then the ethylene diffuses back out of the nodule.

PAGE 55

49 The nodule system to be assayed is assmed to be intact and in a flow-through chamber. The chamber is also assiamed to have perfect mixing and a small time constant (ratio of the gaseous volume of the chamber to the volumetric flow rate through the chamber). If at time zero acetylene is added to the air supply to a final concentration of 3 -3 0.10 mm mm there will be a lag period before steady-state ethylene production is reached. This lag should be made up of 5 basic components: 1) the time required for the chamber to reach the final acetylene concentration, 2) the lag required for acetylene to diffuse into the nodules, 3) the time required for the ethylene concentration inside the nodules to reach its final value, 4) the time required for ethylene to diffuse out of the nodules, and 5) the time required for the ethylene concentration in the chamber to reach steady-state. The first and fifth component will depend on the assay chamber volume and the flow rate of the gas supply through the chamber. To minimize the effects of these two components the time constant for the assay chamber used in this analysis and in the following experiments was only two seconds as a result of the rapid volimetric flow rate through an assay chamber of relatively small volume. Relating Time-To-Steady-State with the Nodule Gas Permeability Because the time required to reach steady-state is a function of the permeability of the diffusion barrier and the nodule volimie, and because the ethylene production rate is governed by Michael is-Menten kinetics, there is no direct relationship between these parameters which lends itself to an analytical solution. Thus, in order to relate P with

PAGE 56

50 the time to steadystate, numerical methods must be employed. To do this a series of nodules were simulated in order to generate a table of predicted times-to-steady-state for nodules of various dimensions and with different values of P. Simulated nodules were assimied to be spherical and to have outer diameters of 2.0, 2.5 and 3.0 mm. The corresponding diameters of the nitrogen fixing inner zone for each of these nodule sizes were calculated according to Bergersen (1982) as being 1.48, 1.84 and 2.38 mm, respectively. Thicknesses used for the diffusion barrier were 10, 20, 30, 40, 50, 60 and 70 micrometers, and the value of the diffusivity of acetylene and ethylene was calculated from published values for diffusivity in plant tissues (Berry and Norris, 1949; Dungey and Pinfield, 1980) which were extrapolated to 26C (Carlson, 1911) as -4 2 -1 4.6*10 mm s The solubility of acetylene was assumed to be l.C and that of ethylene to be 0.108 (Orcutt and Seevers, 1937). These assumed -3 values resulted in simulated values of P which ranged from 6.57*10 to 4.60*10"^ mm s'"*". If the time constant of the assay chamber is known, and if perfect mixing occurs in the chamber then the acetylene concentration at the outer surface of the nodule can be modeled as V = ^fin (1 (4.5) where A^^ is the concentration of acetylene in the assay chamber and at the outer most surface of the diffusion barrier (mm^ mm""^), A^ is the fin final concentration of acetylene in the chamber (mm^ mm~"^) after mixing

PAGE 57

51 3 -3 has been completed and is assumed to be 0.10 mm mm t is the time elapsed since acetylene was introduced to the chamber (s), and tau is the chamber time constant assumed to be 2 sec. Since the assumed thickness of the diffusion barrier (10 to 70 micrometers) is small compared to the diameter of the sphere (1.48 tc 2.38 mm), diffusion across the barrier can be modeled with slab geometry. To do this the barrier was divided into compartments 1.25 micrometers in width and the diffusion of acetylene and ethylene across these compartments was modeled with a continuity equation such that dc/dt = (flow. flow ^ + generation) / vol. (4.6) in out 1 where dc/dt is the change in concentration of the diffusing gas with 3 -3 -1 time (mm mm s ), flow, and flov; ^ are the flow rate of the gas in out 3 -1 into and out of the compartment respectively (mm s ) generation is the rate of generation of the gas in the compartment (mm"^ s ) and vol^ 3 is the compartmental volume (mm ) Since ethylene is not generated in the diffusion barrier. Equation 4.6 can be used to describe the inward diffusion of acetylene such that dCa./dt = D (Ca.^, 2*Ca. + Ca. ) / (1.25 10 (4.7) 1 1+1 1 1-1 3 -3 where Ca^ is the concentration of acetylene (mm mm ) in the ith compartment (with i=0 at the outer edge of the barrier), and D is the 2 -1 diffusivity (mm. s ) of acetylene in the compartment. Once the

PAGE 58

52 acetylene reaches the inner surface of the barrier it enters the inner volume of the sphere where it is consiimed such that R = (Vol Vm Ca^) / (Km + Ca^) (4.8) 3 -1 where R is the rate of ethylene production (mm s ), Vol is the volume 3 of the sphere inside the diffusion barrier (mm ), Vm is the maximum rate -A of ethylene production per unit volume (initially assumed to be 6.5*10 3 -3 -1 mm. mm s ) Ca is the concentration of acetylene in the inner most n 3 -3 compartment of the diffusion barrier (mm mm ), and Km is the Michaelis-Menten constant for reduction of acetylene by nitrogenase 3 -3 ( initially assum.ed to be 0.004 mm mm ) The outward diffusion of ethylene in each compartment was also modeled using the continuity equation and resulted in a series of equations similar to Equation 4.7. The Euler's finite difference method was used on a mini -computer to solve this series of differential equations and the simulation was run until the calculated rate of ethylene leaving the assay chamber was greater than 88% of the predicted final steady-state value which was assumed to be the value of R when Ca n 3 -3 was 0.10 mm mm For each nodule size and barrier thickness which was simulated, the calculated rate of ethylene flow out of the assay chamber was plotted as a function of time. Fig. 4.3 is a series of such

PAGE 59

53 Tinne(seconds) Fig. 4.3. Simulated ethylene flow rate out of the assay chamber (F ) for a 2.0 mm diameter nodule at four different barrier permeabilities

PAGE 60

54 simulated data for 2.0 mm outer diameter nodules at four different barrier permeabilities where P is calculated as P = D S / Lx (4.9) where S is the solubility of acetylene in the cytoplasm (assumed to be 1,0), and Lx is the thickness of the diffusion barrier (mm). Except for the the initial few seconds of the simulation each of these curves appear to be exponential in nature and to fit the general exponential equation F = F,. (1 e"^''^'') (4.10) e f m where F is the ethylene out flow (mm"^ s ) from the chamber at time t e (s), F^. is the final steady-state ethylene out flow rate (nm"^ s fin and tu is the associated time constant (s). Equation 4.10 can be transformed to a linear equation such that ln(l F /F^. ) = -d/tu) t (4.11) e fin where In is the natural logarithm. Eq-uation 4.11 describes a straight line with slope -(1/tu). Thus a simple linear regression can be used to find the time constant (tu) associated with the simulated data sets for each nodule size and permeability. Figure 4.4 is the same data presented in Fig. 4.3 which has been transformed according to Equation 4.11. The data from each of the simulations was transformed in this

PAGE 61

Fig. 4.4. Linear transformation of the simulated ethylene flov; rate out of the assay chamber for a 2.0 mm diameter nodule with four different barrier permeabilities.

PAGE 62

56 manner, and a linear regression run on the data which followed the initial lag period or after t = 10 s. Results from this analysis for the simulated data are presented in Table 4.2 where the estimated time constants for steady-state ethylene production for nodules of three diameters and seven nodule gas permeabilities are presented. Table 4.2. Estimated time constants for simulated nodules Barrier Nodule gas Estimated Time Constants (s) Length Permeability For Nodule Diameters (mm) Of (urn) (mm s 10 ) 2.0 2.5 3,0 10 46.0 6.39 7.74 9,73 20 23.0 11,50 14.26 18,36 30 15.3 17.52 21.63 27.86 40 11.5 23.75 29.26 37.59 50 9.20 30.13 37.13 47.55 60 7.67 36.78 45.13 57.75 70 6.57 43.62 53,40 68.23 In Fig. 4.5 nodule permeability is plotted against the estimated time constants for each of the three nodule sizes reported in Table 4.2. Ideally ethylene flow rates from an assay chamber with intact nodules in it could be measured as a function of time and these data regressed using Equation 4.11 to solve for an associated time constant. The nodules could then be harvested, measured, and the mean nodule diameter determined. Figure 4,5 could then be used to find the average nodule gas permeability for the assayed nodules by extrapolating between the values for the closest nodule sizes simulated.

PAGE 63

•C I I I I • 3" 0 10 30 50 70 Time Constant (seconds) Fig, 4.5. Nodule gas permeability verses the estimated time constants for simulated nodules of three different diameters.

PAGE 64

58 Sensitivity Analysis of the Lag-Phase Model The data used in Fig. 4.5 were derived from simulated data which 1) assumed only one value of the Km, 2) used only one value of Vm, and 3) assumed that nodules of only one size were present in the chamber and that this nodule size was known. The sensitivity of the lag-phase model to errors in these assumptions was tested. The relative sensitivity of P to one of the factors in the model is defined here as PS = (P^-P2)/(F^-F2) F^/Pj^ (4.12) where RS is the relative sensitivity of P to factor F, and P^ and P^ are the predicted values of the nodule gas permeability when the factor of concern is F^ or F^, respectively. Equation 4.12 was used to test the relative sensitivity of P to changes in the nodule diameter, the Km and Vm. Figure 4.5 indicates that for a given time constant, the estimated values of nodule gas permeability vary greatly for nodules of different sizes. Using these data and setting F^ equal to a diameter of 2.5 mm, RS was calculated for a change in diameter of -0.5 and 0.5 mm, at tu values of 14, 26 and 37 s. The relative sensitivity of the model to errors in nodule diameter are given in Table 4.3.

PAGE 65

59 Table 4.3. Relative sensitivity of P to nodule diameter. CHANGE IN DIAl-lETER TIME CONSTANT ( SECONDS ) (mm) 14 26 37 -0.5 0.94 0.89 0.87 0.5 1.19 1.36 1.34 Errors in nodule diameter resulted in substantial errors in the predicted value of P. The relative sensitivities of P to nodule diameter reported in Table 4.3 range from 0.87 to 1.36. This indicates that for given error in nodule diameter, the estimated value of P will also be in error by a factor of 13% less than the error in diameter to 36% greater than the error in diameter. In order to test the relative sensitivity of P to either Km or Vm, a second set of simulations was run. In these simulations two 3 -3 -4 additional values of Km (0.002 and 0.008 mm mm ) and Vm (2*10 and -4 3 -3 -1 8*10 mm mm s ) were assumed. Results of these simulations are presented in Tables 4.4 and 4.5. Table 4.4. Relative sensitivity of P to Km CHANGE IN^Km TIME CONSTANT ( SECONDS ) mm mm 14 26 37 -0.002 0.05 0.05 0.05 0.004 0.03 0.03 0.03

PAGE 66

60 Table 4.5. Relative sensitivity of P to Vm CHANGE Vm^ TIME CONSTANT (s) mm mm s 26 37 -4.5*10 -4 0.01 0.01 1.5*10 -4 0.00 0.01 The relative sensitivities of P to changes in Km and Vm are quite low ranging from 0.03 to 0.05 and 0.0 to 0.01 respectively. This indicates that if either the Km or the Vm of the nodules being assayed differs from the values used to generate Fig. 4.5 by as much as 100%, the error in the predicted nodule gas permeability would be 5% or less. Predicting P for Systems of Mixed Nodule Sizes Davis (1984) demonstrated that the time required for a mixture of nodules of various sizes to reach steady-state ethylene production could be dominated by the larger nodules. The heterogeneous composition of nodule sizes which are found on intact plant roots could therefore introduce an error in the estimations of nodule gas permeability derived by the lag-phase technique. To test whether the lag-phase model could predict the nodule gas permeability of a mixture of nodules of different sizes, simulated data from 2.0, 2.5 and 3.0 mm nodules were combined to form a mixed data set in which there were equal numbers of each of these nodule sizes. This combined data set was transformed using Equation 4.11 and a linear regression used to solve for the associated time constant. The mean nodule diameter for the mixed system was found by calculating the total nodule surface area in the simulated chamber.

PAGE 67

61 dividing this by the number of nodules in the chamber to estimate a mean surface area per nodule, and then calculating the spherical diameter which would result in this surface area. The resultant mean nodule diameter for an equal mixture of nodules from each size group was found to be 2.2 mm. This simulation was repeated for three different values of P which were 0.092, 0.0115, and 0,0153 mm s'-*-. The data in Fig. 4.5 were then used to extrapolate for the estimated P for each of these mixed systems. Table 4,6 gives the resultant percent error in the predicted values of P. Table 4,6. Percent error in P for a mixed nodule system with equal numbers of nodules at 2.0, 2.5, and 3.0 mm. PERMEABILITY PERCENT ERROR IN THE mm s~ PREDICTED PERMEABILITY 0.0092 -12.9 0.0115 -13.7 0.0153 -13,7 Mixing an equal number of 2.0, 2.5, and 3.0 mm diameter nodules does introduce an error into the estimated values of gas permeability. The error is approximately -13% and is consistent across the full range of permeabilities tested. Since it is unlikely that a plant will have an equal number of nodules at each of the sizes that were simulated, nodule size distributions from intact hydroponically grown plants were examined. Nodule data were collected from 19 plants which were used to test the lag-phase technique and which are described below. The nodules were

PAGE 68

62 harvested, individually measured and ranked into three size categories based on the average nodule diameter. Simulated data sets consisting of mixtures of nodules of different sizes were then constructed based on the results of these rankings. Most of the 19 plants had nodule distributions dominated by nodule diameters of less than 3 mm, and with an approximate nodule distribution of 48% at 2 mm, 37% at 2.5 mm and 15% at 3.0 mm. A simulated data set with this distribution was considered to represent an average plant. Two simulated data sets were also constructed which represented the most extreme nodule distributions found among the 19 plants that were measured. These extreme data sets represented a plant dominated by large nodules with an approximate distribution of 18% at 2 mm, 29% at 2.5 mm and 53 % at 3 mm, and one plant dominated by small nodules with an approximate distribution of 82% at 2 mm, 16% at 2.5 mm and 2% at 3 mm. Each of these nodule data sets were then simulated with the three values of nodule gas permeability used above. The percent error in the estimated mean nodule gas permeability for each of these mixed nodule data sets are presented in Tables 4.7 through 4.9. Table 4.7. Percent error in P for a representative mixed nodule system with 48, 37, and 15% of the nodules at 2.0, 2.5, and 3.0 mm, respectively. PERMEABILITY mm s '^ PERCENT ERROR IN THE PREDICTED PERMEABILITY 0.0092 9.9 0.0115 9.6 0.0153 -13.1

PAGE 69

63 Table 4.8. Percent error in P for a mixed nodule system dominated by large nodules with 18, 29, and 53% of the nodules at 2.0, 2.5, and 3.0 mm, respectively. PERMEABILITY PERCENT ERROR IN THE mm s""*PREDICTED PERMEABILITY 0.0092 9.1 0.0115 9.6 0.0153 9.8 Table 4.9. Percent error in P for a mixed nodule system dominated by small nodules with 82, 16, and 2% of the nodules at 2.0, 2.5, and 3.0 mm, respectively. PERMEABILITY PERCENT ERROR IN THE mm s PREDICTED PERMEABILITY 0.0092 5.0 0..0115 5.2 0.0153 5.2 Each of the simulations of data sets containing nodules of different sizes resulted in underestimations of the nodule gas permeability. In mixtures which contained equal numbers of nodules at each of the three sizes, the error in the predicted value of P was approximately -13%. In the simulated data set which was representative of the plants which are reported below, the error in P ranged from -9.6 to -13.1%. The error associated with P for mixtures which represented plants dominated by large or small nodules was approximately -9.5 and -5%, respectively.

PAGE 70

64 Predicting P for nodules of mixed sizes with the lag-phase technique resulted in an under-estimation of the correct nodule gas permeability. The magnitude of this error varied from approximately -5% to -14% depending on the nodule size distribution in the mixed sets. This error in the estimated P was relatively small and was very consistent across the range of actual permeabilities simulated for each individual data set. This indicates that while the predicted P for a given plant may be underestimated, this error v;ill be consistent across treatments Thus the lag-phase technique can be used to determine if the nodule gas permeability of a set of nodules changes over time or in response to an experimental treatment. Conclusions from Simulated Data The lag-phase method for determining nodule gas permeability results in estimates which are highly independent of the biochemical characteristics of nitrogenase. This independence from Vm and Km indicate that this method can be used even if these parameters change over the course of an experiment or as part of the experimental treatment. The m.ethod is sensitive to errors in the estimate of the mean nodule diameter of the nodules assayed. This problem can, however, be solved as the nodule sizes can be directly measured. Mixtures of differently sized nodules in an assay chamber may cause errors in the absolute value of the predicted nodule gas permeability. This error, however, will be consistent across experimental treatments which might effect the value of the nodule gas permeability. It is concluded that the lag-phase model should predict the nodule gas permeability with a

PAGE 71

65 high degree of independence from the biochemical characteristics of nitrogenase activity as long as the mean diameter of the nodules being assayed is known. Validation of the Lag-Phase Model with Intact Plant Material Nineteen intact hydroponically grown soybean plants were individually assayed under ambient oxygen and at 26C to generate data to test the lag-phase model. Plant material, growth conditions, and the assay chamber were identical to those described in Chapter III. The night before a plant was to be assayed it was transferred from a hydroponic-growth chamber to the flow-through assay chamber. The following morning the air flow rate to the chamber was increased from 0.33 to 10.0 mL s""*". This resulted in a time constant for the assay chamber of 2 s. Acetylene was added to the supply gas just upstream from the assay chamber to a final concentration of 0.10 mm"^ mm ^. Onemilliliter gas samples were drawn from the gas exiting the chamber every 4 s until 128 s after the addition of acetylene at which time the acetylene was removed from the gas supply, the chamber flushed with air and the flow rate returned to 0.33 mL s ^. The gas samples were then injected into a gas chromatograph fitted with a flame ionization detector to determine the ethylene flow rate out of the chamber at the time the individual samples were drawn. The acetylene concentration in the assay chamber and the ethylene production rate reached steady-state in less than 8 s and in approximately 90 s, respectively. Since the assay was complete in less than 3 min, the inhibitory effects of acetylene on nitrogenase activity were avoided (see Fig. 3.3) and the

PAGE 72

66 final steady-state rates of acetylene-reduction were assumed to represent maximum nitrogenase activity. In order to calculate nodule gas permeability the mean nodule diameter for the nodules in the assay chamber was needed. This was calculated by dividing the estimated total nodule surface area in the chamber by the number of nodules to obtain a mean nodule surface area from which an average diameter was calculated as if the nodules were spherical. Total nodule surface area was calculated by the method previously described by Weisz et al (1985). In brief, 139 nodules were harvested from hydroponically grown plants similar to those used in these experiments. The length (defined as the longest nodule dimension), width and height of each nodule was measured to the nearest half millimeter and the nodule surface area calculated assuming the nodule was ellipsoidal and using a numerical solution to the equation for the surface area of a quadratic surface. These data for nodule surface area v;ere then regressed against the longest nodule dimension. 2 A quadratic model gave the best fit (R = 0.98) and is presented in Equation 4.13. Narea = 3.008 L + 1.210 (4.13) 2 Narea is the nodule surface area (mm ) and L is the longest nodule dimension (mm) To calculate the total nodule surface area for an experimental plant the nodules were harvested, L was determined for each nodule and Narea calculated according to Equation 4.13. The total nodule surface area assumed to be the sum of the estimated surface areas from each nodule.

PAGE 73

67 Once the mean nodule diameter was determined the non-steady-state ethylene production rate data were transformed using Equation 4.11 and a linear regression used to solve for an associated time constant. The data in Fig. 4.5 were then used to extrapolate for the mean nodule gas permeability of the assayed nodules. Non-steadystate ethylene production data are presented for a representative plant in Fig. 4.6 where ethylene production as a percent of the steady-state rate and the linearized transformation of these rates are both plotted against time since acetylene was introduced into 2 the gas supply. The resultant R for the linearized data was 0.99 and the associated time constant was 24.0 s. The mean nodule diameter found for these nodules was 2.5 mm and the predicted nodule gas permeability -3 -1 was 13.9*10 mm s The mean nodule gas permeability for the 19 plants assayed in this fashion was found to be 13.3*10 mm s with a standard error of the -3 -1 mean of 0.61*10 mm s These permeabilities are in close agreement -3 with the permeabilities for well -watered field-grown soybeans (5.0*10 -3 -1 to 10*10 mm s ) reported by Weisz et al (1985). Assuming a diffusivity of acetylene through the nodule cortical barrier of 4.6*10 ^ 2-1 mm s as used above, the resultant mean thickness of the diffusion barrier can be calculated to be 30.5 um. This barrier thickness is very similar to that calculated by Sinclair and Goudriaan (1981), who estimated that a barrier of 45 um would be required to reduce the oxygen concentration inside a soybean nodule to values low enough to permit nitrogenase to function. Sheehy et al. (1983) using the respiration model described above reported somewhat lower nodule gas permeabilities

PAGE 74

68 Fig. 4.6. Ethylene production rate as percent of the final steady-state value, and the linear transformation of ethylene production rate verses time in seconds since acetylene was introduced into the assay chamber.

PAGE 75

69 for white clover nodules at ambient oxygen concentration in the range of 1.87*10"^ to 2.87*10"^ mm s"-"". Conclusions Two methods for estimating the nodule gas permeability were reviewed and found to be unreliable. A new method based on non-steadystate ethylene production of intact nodules after exposure to acetylene was developed and tested. This technique is rapid, requiring less than three minutes to complete. It gives estimates of both nodule gas permeability and the maximum rate of acetylene-reduction and it is independent of assumptions regarding nodule respiration or the Km for nitrogenase. Finally, the lag-phase model was tested with intact hydroponically grown soybean plants and the resultant values of nodule gas permeability were very close to both theoretical estimates and predictions made using the other techniques. It is concluded that this method for estimating nodule gas permeability is suitable for testing the hypothesis that altered oxygen concentrations result in a regulation in the permeability of the diffusion barrier in the nodule cortex.

PAGE 76

CHAPTER V NODULE GAS PERMEABILITY, RESPIRATION AND NITROGEN FIXATION IN RESPONSE TO ALTERED OXYGEN CONCENTRATION In Chapter III it was concluded that in order to maintain a stable and low internal nodule oxygen environment after altering the rhizosphere oxygen concentration the nodule respiration or the gas permeability would have to adjust. The data presented in Chapter III, however, did not support the hypothesis that the adaptive mechanism involved a change in nodule respiration. Sheehy et al (1983) reported for white clover nodules that the gas permeability of the diffusion barrier might be sensitive to the oxygen gradient across it and vary in magnitude in response to altered external oxygen concentrations. Their hypothesis predicts that altered oxygen concentrations around the nodules will effect the nodule activity, but that the permeability of the nodule cortical diffusion barrier will change over time such that the flux of oxygen into the nodule will return to rates similar to those originally observed under ambient oxygen conditions. This implies that the gas permeability of the nodules may be under physiological control and that by changing this parameter nitrogen fixation rates might be regulated. The purpose of this research was to determine if the nodule gas permeability is responsive to the rhizosphere oxygen concentration, and if so, whether it could be the mechanism responsible for regulating nodule activity in response to alterations in the oxygen environment. 70

PAGE 77

71 Materials and Methods The intact hydroponic soybean plants used in these experiments were grown under the same conditions as described in Chapter III. Individual intact plants were transferred from the hydroponic growth chambers to a flow-through stainless steel assay chamber at the start of an experiment. The mean nodule gas permeability of the nodules in the chamber was assayed using the lag-phase technique described in Chapter IV. To do this the air flow rate through the assay chamber was increased from 0.33 mL s to 10 mL s ^. Acetylene was introduced to the air supply just upstream from the chamber to a final concentration 3 -3 of 0.10 mm mm and 1-mL gas samples were drawn from the gas exiting the chamber every 4 s for approximately 128 s. At that time, the acetylene was removed from the gas supply, the chamber flushed with air and the flow rate returned to the normal rate of 0.33 mL s ^. The gas samples were injected into a gas chromatograph fitted with a flame ionization detector in order to determine the non-steady-state ethylene production rate of the nodules in the chamber at the time the samples were collected. An associated time constant for the non-steady-state response data was determined from which the nodule gas permeability was calculated using Fig. 4.5. The mean nodule diameter of the nodules in the assay chamber was needed for this calculation and was determined after harvesting the nodules at the end of the expermiment. Mitrogenase activity was assayed as part of the lag-phase technique used for determining the nodule gas permeability. Steadystate acetylene-reduction was typically reached in 60-s to 90-s. Since this assay procedure was completed well in advance of the exposure time

PAGE 78

72 nescessary for acetylene to inhibit nodule activity (see Chapter III) the steady-state acetylene-reduction rate was assumed to represent nitrogenase activity. Nodule respiration rate was also determined on about half the plants as described in Chapter III. In brief, a Walker-type Clark-style oxygen electrode was used to determine the differance between the oxygen concentration in the gas stream supplied to, and exiting from the assay chamber. This difference in oxygen concentration resulted from root plus nodule respiration. Root respiration rate at ambient oxygen and at the final oxygen concentration to which the plant was exposed, V7as assayed at the end of each experiment after harvesting the nodules and returning the plant to the assay chamber. Nodule respiration rate v;as calculated as the difference between total root plus nodule oxygen uptake and bare root respiration rate. Experimental Design The night before a plant was to be assayed it was transferred from a hydroponic growth chamber to the assay chamber which was maintained at -2 -1 26C and provided with 950 to 1400 uE m s of photosynthetically active radiation depending on position in the canopy. The following morning nodule activity and gas permeability were assayed at ambient oxygen. The oxygen concentration in the gas supply was then altered to 3 -3 0.1 mm mm (for 10 plants) and these parameters repeatedly assayed \intil steady-state was reached. Three plants to act as a control treatment were repeatedly assayed over a 30-h period without switching the oxygen concentration from ambient levels.

PAGE 79

73 For exposure to supra-ambient oxygen concentration a 30-L mixing tank was introduced into the air supply line upstream of the assay chamber. This increased the time used for a step change in oxygen concentration from the normal 5 min to over 1 h. This was done in an attempt to avoid nitrogenase oxygen inactivation as may have been observed in the experiments described in Qiapter III. Nodule gas permeability and acetylene-reduction for four plants initially assayed 3 -3 at ambient oxygen and then switched to 0.28 mm mm were repeatedly 3 -3 assayed. After about 2 h at 0.28 mm mm oxygen the system v;as switched to 0.32 mm'^ mm ^ oxygen and the plants assayed until steadystate was obtained. Results Control Plants Nodule gas permeability, acetylene-reduction and nodule respiration for three plants were repeatedly assayed under ambiient conditions for over 30 h. Data from these control plants as a percent of the initial value are plotted in Fig. 5.1 against time since the growth chamber lights came on the first morning of the experiment. No diurnal trends were apparent in these data as nodule gas permeability, acetylenereduction and nodule respiration were very stable over this 30 h light, dark and light period. 3 -3 Study at 0.1 mm mm Oxygen Nodule gas permeability, acetylene-reduction and nodule respiration •3 for two representative plants switched from ambient to 0.1 mm mm

PAGE 80

74 Fig. 5.1. Nodule gas permeability (open circles), maximum acetylenereduction (solid squares), and nodule respiration (dots) as percent of the initial value verses time since the growth chamber lights came on for three control plants assayed only under ambient oxygen conditions.

PAGE 81

75 oxygen are presented in Figure 5.2, The effect of reduced oxygen on nodule respiration and acetylene-reduction is similar to that reported for plants in Chapter III. In contrast to these two parameters, nodule gas permeability began to increase when the oxygen concentration was decreased and continued to increase to values well above those observed under ambient conditions. A summary table of the final steady-state values for nodule gas permeability and acetylene-reduction as percent of the initial values 3 — 3 for all plants assayed at 0.1 mm mm oxygen is presented in Table 5.1. Steady-state rates of acetylene-reduction were similar to those observed at ambient oxygen. Nodule gas permeability however, increased by 63% over the initial values observed under ambient conditions. Table 5.1. Steady state values at 0.1 mm mm oxygen as percent initial value at ambient oxygen. Expt. # % Acetylene Reduction % Nodule Gas Permeability 1 89 156 2 119 158 3 110 145 4 107 307 5 109 129 6 111 153 7 106 145 8 95 146 9 116 157 10 104 138 MeaniS .E: 106. 62. 9 163.416.2

PAGE 82

76 Time (hours) Fig. 5.2. Nodule gas permeability (open circles), maximum acetylenereduction (open squares) and nodule respiration (dots) as percent of the initial value verses time since the growth chaiTiber lights came on for an two plants assayed under ambient conditions and then switched to 0.1 mm mm oxygen. Arrows indicate the time when the oxygen concentration was altered.

PAGE 83

77 Study at 0.28 and 0.32 mm mm Oxygen Data for nodule gas permeability, acetylene-reduction and nodule respiration for one of the four plants assayed at supra-ambient oxygen concentrations are presented in Fig. 5.3. Nodule respiration initially decreased with elevated oxygen concentration, then increased to values approximately 20% above ambient rates and finally declined to values similar to those initially observed under ambient conditions. Acetylene reduction rate also decreased with elevated oxygen concentration and then recovered to nearly ambient values. Nodule gas permeability decreased with elevated oxygen concentration and reached final values 34% below the initial levels. Partial oxygen inactivation of nitrogenase may have occurred in two of the four plants as nodule respiration rate and acetylene-reduction rate initially decreased after elevating the chamber oxygen concentration. The remaining two plants showed little or no short-term effect of oxygen concentration on these parameters. In all four plants elevated oxygen concentration resulted in decreased nodule gas permeability. A sxmmary of the steady-state values of acetylenereduction rate and nodule gas permeability as percent of the initial values for these four plants assayed at supra-ambient oxygen concentration is presented in Table 5.2. Steady-state values of 3 -3 acetylene-reduction rate at 0.32 mm mm oxygen were similar to those observed under ambient oxygen. The mean nodule gas permeability, however, decreased by approximately 32%,

PAGE 84

78 X I40r5> — — I — — — — — ^ — — — — — — ^ o i I 1 1 J 1 1 1 L-T 0 10 14 18 22 26 30 Time (hours) Fig. 5.3. Nodule gas permeability (open circles), maximum acetylenereduction (open squares) and nodule respiration (dots) as percent of the initial value verses time since the growth chamber lights came on for an individual plant assayed^unde^ ambient conditions and then switched to 0.28 and then to 0.32 inm mm oxygen. Arrows indicate the times when the oxygen concentrations were altered.

PAGE 85

79 Table 5.2, Steady state values at supra ambient oxygen as percent initial value at ambient oxygen. Expt. % Acetylene # Reduction % Nodule Gas Permeability 1 2 3 4 106 100 91 92 60 69 76 66 MeantS.E: 97.33.5 67.83.3 Discussion The purpose of this research was to test the hypothesis that the nodule gas permeability is sensitive to the external oxygen concentration and that with alterations in the oxygen environment it adjusts in order to maintain a constant flux of oxygen across the nodule cortical diffusion barrier. This hypothesis predicts that at subambient oxygen concentrations where the oxygen gradient across the nodule cortex is reduced, the nodule gas permeability will increase to allow a greater flux of oxygen into the nodule, and conversely under supra-ambient oxygen concentrations the nodule gas permeability should be decreased. The data on nodule gas permeability from these experiments support this hypothesis. The nodule gas permeability began to increase as soon as the oxygen concentration around the hydroponically grown plants was reduced (Fig. 5.2) and at steady -state had increased approximately 63% over the ambient values. Conversely, nodule gas permeability began to decrease when the oxygen gradient was elevated (Fig, 5.3) and final steady-state values were about 32%

PAGE 86

80 decreased over ambient values. Nodule gas permeability did respond to the external oxygen concentration around the nodules of these hydroponically grown soybean plants. Sheehy et al. (1983) concluded that a similar mechanism for adjusting the nodule gas permeability to altered oxygen concentrations existed in white clover nodules. Sheehy et al (1983) reported that the time required for white clover nodules to reach new steady-state respiration (carbon dioxide evolution) rates after an increase in the external oxygen concentration was approximately 45 min. The final rates for respiration and nitrogenase activity reported by these authors was not constant across the range of oxygen concentrations used as 3 -3 respiration at 0.3 mm mm oxygen was 127% higher than that observed at 0.05 mm^ mm"'^ oxygen. Witty et al (1984) reported similar findings for pea and lucerne but were unable to detect an adjustment in nodule gas permeability in soybean or sainfoin. Witty et al. (1985) were also unable to detect an adjustment mechanism in soybean. These findings led Witty et al. (1984) to hypothesis that there are two groups of legxnnes. In the first group nodules were assumed to be capable of rapid adjustment in nodule gas permeability and this would explain the results of these authors with white clover and pea nodules. The second group of nodules was assumed to have either no, or only a very slow adjustment capability. Soybean was placed in this second group. The experminents of Witty et al (1984, 1985), however, were only concerned with the short-term response (1 h or less) of the nodules to altered oxygen concentrations and this might explain why a change in nodule gas permeability was not indicated for soybean.

PAGE 87

81 The data presented here indicate that soybean nodules are capable of regulating the nodule gas permeability and that the time required for this regulation is longer than that reported by Sheehy et al (1983) for white clover. It is possible that the regulator mechanism in soybean is slower than that of white clover, but these data also indicate that it may be more dynamic as complete adjustment of nodule activity v/as possible over a wide range of external oxygen concentrations (0.1 to 0.4 mm mm ) It is also possible that the legumes assayed by Sheehy et al. (1983) and Witty et al. (1984, 1985) would also have completed full recovery of nodule activity over these oxygen concentrations if the experimental procedures used by these authors had been extended for longer time periods. Nodules are apparently capable of regulating nitrogen fixation rates by changing the rate at which oxygen can diffuse into the nodule interior. Oxygen diffusion into soybean nodules is initially dominated by diffusion in air spaces in lenticel like structures in the outer cortex (Pankhurst and Sprent 1975a). Interior to the outer cortex there is a layer of cells in the inner cortex which lack intercellular air spaces and which has been hypothesized to be the site of the oxygen diffusion barrier (Tjepkema and Yocum, 1974; Sinclair and Goudriaan, 1981). Pankhurst and Sprent (1975a) reported that when subjected to drought -stress the air spaces in the lenticels close as the cells in the outer cortex lose turgor, and this led Pankhurst and Sprent (1975b) to hypothesize that a decrease in the nodule gas permeability would be associated v/ith drought. This hypothesis was supported by the data of Weisz et al (1985) which indicated that drought-stressed soybean

PAGE 88

82 nodules had significantly lower values of nodule gas permeability than did nodules on well-watered control plants. Turgor pressure in the lenticular cells seemingly affects the air spaces in the outer cortex and the magnitude of the nodule gas permeability. It is possible that unknown mechanisms which regulate turgor in the cortical cells represent the basis of the nodule capacity for regulating the nodule gas permeability and subsequently nitrogen fixation.

PAGE 89

CHAPTER VI SOIL TEMPERATURE, NODULE GAS PERMEABILITY AND DIURNAL CYCLES IN SOYBEAN NITROGEN FIXATION RATE Recent research has indicated that diurnal cycles in symbiotic nitrogen fixation rate in intact leguminous plants may be the result of temperature changes around the plant roots and nodules. Eckart and Raguse (1980) reported that diurnal cycles in nodule acetylene-reduction rates in growth chamber grown subterranean clover ( Trifolium subterraneum L.) were related to the soil temperature. Similarly, Schweitzer and Harper (1980) reported that when maintained at constant root temperature diurnal light cycles did not effect the nodule activity of growth cabinet grown soybean plants ( Glycine max L.). Diurnal fluctuations in nodule activity were only evident when the temperature around these roots and nodules was altered. Winship and Tjepkema (1983, 1985) also found that acetylene-reduction rates for intact green housegrown Alnus rubra root nodules were related to nodule temperature. Further evidence that diurnal cycles in nitrogen fixation rate are related to changes in soil temperature has been reported in studies using intact field-grown soybean. Denison and Sinclair (1985) found that there was a high correlation between in situ soybean nodule activity and soil temperature measured at a depth of 0.1 m. Similarly, Sinclair and Weisz (1985) found that there was a linear relationship between soil temperature and acetylene-reduction by intact field-grown soybean nodules at temperatures below 30C. In both these studies diurnal variability in nitrogen fixation rate was correlated v/ith changing soil temperature. 83

PAGE 90

84 While there is evidence which indicates that diurnal cycles in soil temperature are the environmental cause of daily fluctuations in nodule activity, it is still frequently assumed that diurnal variability in nitrogen fixation is the result of changing levels of photosynthetically active radiation coupled with a rate limiting nodule photosynthate supply (Hardy and Havelka, 1976). The studies of Denison and Sinclair (1985) and Sinclair and Weisz (1985) with field-grown soybean do not support this hypothesis, but their results are based on correlational data and the natural diurnal cycles of photosynthetically active radiation and soil temperature were not experimentally manipulated in a manner which would clearly separate their individual effects. Root temperature and photosynthetically active radiation were experimentally manipulated in the experiments cited above by Eckart and Raguse (1980) and Schweitzer and Harper (1980) but as growth chamber plant materials were used their results may not reflect mechanisms which function under field conditions. The initial purpose of this research was to confirm that under field conditions diurnal cycles in soybean nitrogen fixation rate follow the soil temperature and are independent of daily light cycles. To do this, the effects of the light and soil temperature cycles on nitrogen fixation rate were separated by manipulating the phase and amplitude of the temperature diurnal. The results indicated that both the phase and magnitude of cycles in daily soybean nitrogen fixation were independent of fluctuations in photosynthetically active radiation but strictly followed the soil temperature.

PAGE 91

85 Denison et al (in press) postulated that in leguminous systems the flux of oxygen diffusing into nodules acts as the primary limitation to nodule respiration and therefore energy production necessary for nitrogen fixation. This hypothesis predicts that in order for fixation rates to increase in response to elevated temperature the flux of oxygen into the nodules must also be responsive to nodule temperature. Winship and Tjepkema (1983) and Sinclair and Weisz (1985) reported that the nodule gas permeability changed proportionally with temperature in nodules from Alnus rubra and soybean, respectively. Such a change in the nodule gas permeability would affect the rate of oxygen diffusion into the nodule and might explain the elevated rates of nitrogen fixation observed at increasing temperature. The analytical procedures used by Winship and Tjepkema (1983) and Sinclair and Weisz (1985), however, were criticized in Chapter IV where it was dem.onstrated that the resultant values of nodule permeability derived using these methods may not be reliable. Therefore, the second objective of this research was to use the lag-phase technique described in Chapter IV to determine if the gas permeability of soybean nodules is responsive to temperature and may act to regulate nitrogen fixation rates in response to diurnal cycles in soil temperature. Materials and Methods Field Study with Intact Soybean Plants The cultivar 'Biloxi' was grown in Gainesville, Florida, on Arredondo fine sand (loamy, siliceous, hyperthermic Grossarenic Paleudult ) Biloxi is an indeterminate, maturity group VIII cultivar

PAGE 92

86 which was at growth stage Rl (Fehr et al., 1971) when this experiment was conducted from 27 June to 3 July, 1984. Field preparation included application of 550 kg ha ^ of 0-10-20 (N-P_o^-K_0) fertilizer and the incorporation of 3 L ha of trifluralin (alpha, alpha, alpha, trif luoro-2,6-dinitro-N,N-dipropYl-p-toluidine) herbicide. On 2 April, rows spaced 0.9 m apart were seeded with 30 seeds m~^. Open-ended root chambers used for the acetylene-reduction assay (Denison et al., 1983a) were installed in the rows at 1-m. intervals immediately after seeding. Alachlor ( 2-chloro-2 -6 -diethylN-(methyoxymethyl) acetanilide) and chlorpyrifos (0,0-diethyl-0-(3, 5,etrichloro-2-pyridyl ) phosphorothioate ) were then both applied to the plot at a rate of 4 L ha ^. Sprinkler irrigation was applied to the plot to assure well-watered conditions. The crop canopy was closed by the time this experiment was conducted. The assay chambers were modified to allow heating or cooling of the soil inside the chambers. This was done by wrapping a 2-m length of 10mm diameter copper tubing around the chambers before they were buried in the field. During the experiment, water from a controlled water bath was circulated through the copper tubing to regulate the chamber temperature. The chamber temperature was measured with a thermistor (#44032, Omega Engineering Inc.) that had been potted inside a 2-mm diameter aluminum tube and buried at a depth of 70 mm near the center of the chamber. The electrical resistance of the thermistor was monitored by a computer v;hich also regulated the temperature of the circulating water.

PAGE 93

87 The temperature of the soil around four plants was not altered in order to study acetylene-reduction rates under ambient temperature conditions. A temperature-regulated treatment, composed of six plants, was imposed by altering the diurnal soil temperature cycles. To do this the computer controlling the chamber temperature was programmed to expose the plants to a sinusoidal diurnal variation in soil temperature from 19 to 29C. From 27 to 29 June, this experimental temperature cycle was in phase with the daily light cycle with maximum and minimuiri temperatures being reached at noon and mid-night, respectively. In the evening of 29 June, the phase of the temperature cycle was shifted by 12 h such that the soil temperature in the chambers was 29C at mid-night and 19 "C at mid -day. In situ ethylene production rates of intact plants in the openended assay chambers were measured at four acetylene concentrations (0.001, 0.004, 0.008, and 0.01 mm^ mm~"^) at 0600, 1200, 1800 and 2400 EST each day throughout the experiment. The analytical technique described by Denison and Sinclair (1985) was used to calculate the maximum rate of acetylene-reduction at saturating acetylene concentrations. To minimize variability among plants which differed in nodule mass, acetylene-reduction rates were converted to a percentage of the rate observed for a given plant at 1800 h on the second day of the experiment Hydroponic Study with Intact Soybean Plants To be consistent with the field experiments the cultivar 'Biloxi' was used for these hydroponic studies. Seeds were surfaced sterilized

PAGE 94

88 with 2% sodium hypocholorite and germinated on moist filter paper. After germination the seedlings were transferred to growth pouches until the primary root was approximately 50 mm long. The seedlings were then individually transferred to bored #3 rubber stoppers which were placed in the lid of a 1.5 L hydroponic chamber made from 102-mm diameter PVC pipe (described previously in Chapter III) and inoculated with a conmercial Bradyrhizobium japonicum inoculum (Nitragin Corporation). Plants were maintained in half -strength nitrogen-free nutrient solution (Imsande and Ralston, 1981) which was continuously aerated by passing 33.3 mL s"''" of air through an aquarium glass bead bubbler in the bottom of each growth chamber. The growth chambers were submerged in a water bath which maintained the temperature around the roots and nodules at 26"C. Illumination was provided by a "Sun-Brella" (Environmental Growth Chambers, Chagrin Falls, Ohio 44022) which consisted of a multi-vapor metal halide lamp (General Electric #E-37) in combination with one highpressure sodium lamp (General Electric #E-18) in a water-cooled jacket -2 -1 and which provided 950 to 1400 uE m s of photosynthetically active radiation depending on position in the canopy. The photoperiod was adjusted to a 16-h day to assure that plants remained vegatetive through-out the assay period. The day before an individual plant was to be assayed it was transferred from the hydroponic growth chamber to a stainless steel flow-through assay chamber (previously described in Chapter III) An intact plant, including the rubber stopper through which the stem grew in the hydroponic chamber, v/as used by inserting the nodulated root system through the bore of the assay chamber until the rubber stopper

PAGE 95

89 sealed the top of the chamber. An air tight seal at the lower opening of the chamber was made by impregnating the roots in silicon grease (Dow Corning) inside of a split and bored #3 rubber stopper. The assay chamber was then placed on top of a modified hydroponic chamber. Moisturized air was continuously passed through the assay chamber at a rate of 0.33 mL s to assure that the nodules were well aerated. The assay chamber was maintained in the same water bath and under the same lighting conditions as the hydroponic chambers described above. Acetylene reduction and nodule gas permeability for the intact nodules in the assay chamber were assayed as previously described in Chapter IV. In brief, the air flov; rate to the assay chamber was increased from 0.33 to 10 mL s ^. Acetylene was added to the gas supply just upstream from the assay chamber to a final concentration of 0.10 3 -3 mm mm One-milliliter gas samples were collected from the gas exiting the chamber every 4 s until 128 s after the addition of acetylene at which time the acetylene was removed from the gas supply, the chamber flushed with air and the flow rate returned to 0.33 mL s ^. The gas samples were then injected into a gas chromatograph fitted with a flame ionization detector to determine the ethylene flow rate out of the chamber at the time the individual samples were drawn. The final steady-state rates of acetylene-reduction (typically reached in 60 to 90 s after the addition of acetylene to the gas supply) were assumed to represent maximxim nitrogenase activity. The non-steady-state ethylene production rate data were then analyzed using the lag-phase technique described in Chapter IV to calculate the nodule gas permeability for the nodules in the assay chamber.

PAGE 96

90 The night before a plant was to be assayed, the temperature of the water bath was increased to 28C. The following morning at 0900 EST, acetylene-reduction rate and nodule gas permeability were assayed and the temperature of the water bath was lowered to 24 "C. The change in temperature took about 2 h and at noon the assay was repeated. The temperature was then lowered a second time to 20 "C and acetylenereduction and nodule gas permeability assayed again at 1500 EST. This was repeated for a total of four plants. A control treatment consisting of three plants (previously described in Chapter V) was assayed at a constant root temperature of 26 C over a 30 h light, dark and light period. Results Diurnal Field Studies Soil temperature of the ambient and temperature-regulated plants is plotted against time in Fig. 6.1A. Ambient diurnal variability in soil temperature was small, being approximately 2C or less. The gradual and slight decrease in mean ambient soil temperature from day 2 through 4 may have been the result of overcast and rainy weather over that 48-h period. Diurnal variation of soil temperature in the temperatureregulated chambers averaged just under 10C, ranging from approximately 19 to 29 "C. On the evening of the third day the temperature of these chairibers was phase shifted by 12 h. The mean acetylene-reduction rates for each treatment as a percent of that observed at 1800 h on the second day of the experiment are presented in Fig. 6. IB. On each day of the experiment except day four.

PAGE 97

91 1 2 3 4 5 6 Day Fig. 6,1. A) Chaniber temperature verses day for control (O), and temperature-regulated chambers (solid squares). B) Treatment mean acetylene-reduction rates as a percent of those observed at 1800 EST on day 2 verses time since the start of the experiment

PAGE 98

92 the control plants at ambient soil temperature displayed very slight diurnal variability in acetylene-reduction rates with the daily minimum rate being 90% or more of the diurnal maximum. There was a slight decrease in these ambient daily rates from day 2 through day 4. These diurnal cycles in acetylene-reduction rate were in phase with both the soil temperature and the daily light cycles. In contrast to the plants at ambient soil temperature, the temperature-regulated plants displayed large diurnal fluctuations in acetylene-reduction rate. These diurnal trends changed phase by 12 h when the temperature diurnal in the chairibers was shifted on day 3. Acetylene reduction rates for individual plants in the temperatureregulated treatment for the 24 -h period previous to and immediately after the temperature diurnal was shifted on day 3 are plotted against 2 soil temperature in Fig. 6.2. There is a good linear fit, with R equal to 0.85. Temperature and Light Study with Hydroponic Plants Plants assayed at a constant temperature of 26C over a 30-h light, dark and light period did not have diurnal fluctuations in nodule gas permeability or acetylene-reduction rate (see Fig. 5.1). Acetylenereduction rate and nodule gas permeability are plotted against chamber temperature for each plant assayed in Figs. 6.3A and Figure 6.3B, respectively. Under constant light conditions acetylene-reduction rate and nodule gas permeability decreased as the temperature around the nodules was decreased from 28 to 20^0, Acetylene-reduction rate is plotted against nodule gas permeability in Fig. 6.4. There was a close

PAGE 99

93 Fig. 6.2. Normalized acetylene-reduction rate verses soil temperature for individual plants in the temperature-regulated treatment from 1800 EST on day 1 until 1800 EST on day 3.

PAGE 100

Fig. 6,3. A) Acetylene-reduction rate verses chamber temperature (circles) for four individual hydroponicall y grown plants assayed at three temperatures. B) Nodule gas permeability verses temperature (solid squares) for the same hydroponically grown plants assayed at three temperatures.

PAGE 101

95 Nodule Gas Permeability (mnr/s) x10 Fig. 6.4. Acetylene reduction verses nodule gas permeability for four hydroponically grown plants assayed at 28, 24, and 20 ''C.

PAGE 102

96 2 linear relationship between these two parameters with the resultant R equal to 0.90. Discussion The initial objective of this research v/as to confirm that diurnal fluctuations in nitrogen fixation in field-grown soybean are regulated by the soil temperature. Data from the field experiment supported this hypothesis. Control plants which were exposed to minimal daily fluctuations in soil temperature due to the closed canopy had very small diurnal trends in nodule activity and the mean daily rates declined with temperature from day 2 through 4. Over the same time period, plants exposed to a 10C daily variation in soil temperature displayed large diurnal trends in nodule activity. Acetylene reduction rates were always related to the soil temperature, even when the soil temperature was 12 h out of phase with the daily cycle of photosynthetically active radiation. After the third day of the experiment, maximum rates of nodule activity for the temperature-regulated plants occurred at midnight when the chamber temperature was highest. Conversely, minimum rates of acetylenereduction occurred at mid-day when the chamber temperature was only 19C. Thus, nitrogen fixation rates were totally out of phase with shoot photosynthesis. Even after 46 h in this "out of phase" mode, maximum rates of acetylene-reduction which occurred at midnight on day 6 (see Fig, 6. IB) were just as high as had occurred at noon on the first day, or before the phase shift in temperature was initiated. These data indicate that nodule activity in field-grown

PAGE 103

97 soybean is not regulated by current photosynthesis but follows changes in the soil temperature. Similar conclusions can be drawn from the hydroponic plant data. Plants assayed at a constant root temperature had stable nodule activity and gas permeability over a 30-h period which included 8 h in the dark. Conversely, at a constant level of photosynthetically active radiation, nodule activity varied proportionally with root temperature. Variability in acetylene-reduction rate in these hydroponically grown plants was associated with the root temperature. The second objective of this research was to test the hypothesis that variation in soil temperature would effect the nodule gas permeability. The data from the hydroponic plant experiments support this hypothesis. When the temperature in the hydroponic chambers was reduced from 28''C to either 24C or 20C proportional changes in nodule gas permeability occurred simultaneously with the reductions in nodule activity (Fig. 6.4). This is in agreement with the findings of Winship and Tjepkema (1983) and Sinclair and Weisz (1985) who reported a relationship between nodule temperature and gas permeability. Since the thickness of the diffusion barrier in the nodule cortex is small compared to the diameter of the nodule, nodule gas permeability can be defined as P=D*S/Lx (6.1) -1 where F is the nodule gas permeability (mm s ) D is the gas 2 1 diffusivity through the cortical tissue (mm s ) S is the solubility

PAGE 104

98 3 -3 of the gas in the cytoplasm (mm mm ) and Lx is the thickness of the diffusion barrier (mm). With increasing temperature, D and S (for oxygen in water) increase and decrease, respectively. The magnitude of these changes is such, however, that the product of D S will decrease with increased temperature (Altman and Dittmer, 1971). This indicates that if the effects of temperature on P were purely passive, P should decrease at elevated root temperatures, the opposite of what was observed in these data. Assuming that D and S in the cortex react to temperature in a similar fashion as would be expected for aqueous solutions, Equation 6.1 indicates that for P to increase at elevated temperatures Lx must decrease. Such a change in the thickness of the diffusion barrier has been implied from anatomical studies of drought-stressed nodules (Pankhurst and Sprent, 1975a, 1975b). Drought stress has also been shown to effect the nodule gas permeability (VJeisz et al 1985). The work of Pankhurst and Sprent (1975a, 1975b) has led to the hypothesis that turgor in the cells of the nodule cortex may regulate the opening of intercellular air spaces through which oxygen may diffuse into the nodule. If this hypothesis is correct then the nodule gas permeability data from the hydrotionically grown plants reported here indicate that cell turgor in the nodule cortex must be regulated by a process which is sensitive to the soil temperature. This indicates that regulation of the nodule gas permeability in response to soil temperature may not be a purely passive process. In conclusion, nitrogen fixation rates in intact fieldand hydroponically grown soybean root nodules were not effected by diurnal

PAGE 105

99 cycles in photosynthetical ly active radiation, but were strictly related to root temperature. This relationship between root temperature and acetylene-reduction rate was maintained even after 2 d with the light and temperature cycles being completely out of phase. The effects of soil temperature on nitrogen fixation rate may be explained through an effect on the nodule gas permeability. Changes in the nodule gas permeability could in turn regulate the flux of oxygen into the nodule which is necessary for nodule respiration and energy production needed for nitrogen fixation.

PAGE 106

CHAPTER VII CONCLUSION This research was conducted to test the hypothesis that the gas permeability of soybean nodules is variable, possibly under physiological control and can act as a mechanism for regulating nitrogen fixation rates. The data reported in the preceding chapters support this hypothesis. When soybean nodules were exposed to alterations in the rhizosphere oxygen concentration, or in the soil temperature the nodule gas permeability responded to these environmental changes, and appeared to play an important role in determining the nitrogen fixation rates of these nodules. In Chapter II and III the effects of an altered rhizosphere oxygen concentration on intact fieldand hydroponically grown soybean nitrogen fixation rates were explored. The data reported in Chapter III indicated that when nodules were initially exposed to sub-ambient external oxygen concentrations the nitrogen fixation rates changed proportionally. This response of nodule activity to an altered oxygen environment is consistent with the hypothesis that the flux of oxygen entering the nodules is a limiting factor in energy production necessary for nitrogen fixation. The alteration in nodule activity associated with the change in external oxygen concentration was, however, transitory as nitrogenase activity returned to rates which were equivalent to those observed under ambient conditions within several hours. This explains why short-term experiments have consistently 100

PAGE 107

101 indicated that nodule activity is sensitive to the external oxygen concentration while long-term studies have concluded that nodule activity is independent of the oxygen concentration. These data indicated that nodules contain a regulatory mechanism which is sensitive to the oxygen gradient across the cortical diffusion barrier and which can maintain both a stable low internal oxygen environment and constant steady-state nitrogen fixation rates. In Chapter III data were presented which indicated that this adaptive mechanism does not involve an alteration in the nodule respiration rate. In Chapter IV methods for determining the nodule gas permeability which have been reported in the literature were reviewed and each of them found to be unsatisfactory. A new non-destructive method for assaying the mean nodule gas permeability of the nodules on an intact root system was developed and tested using hydroponically grown soybean plants. This analytical procedure was then used to determine if the nodule gas permeability could respond to alterations in the rhizosphere oxygen concentration and these data were reported in Chapter V. It was found that the nodule gas permeability was responsive to the external oxygen concentration. When soybean nodules were exposed to sub-ambient oxygen concentrations the nodule gas permeability increased over a period of several hours. Such an increased nodule gas permeability could explain the recovery in nodule respiration rate and acetylenereduction rate observed in these plants. Conversely, v/hen the rhizosphere oxygen concentration was elevated the nodule gas permeability decreased also over a period of several hours and the final steady-state rates of nodule respiration and acetylene-reduction were

PAGE 108

102 once more very similar to those observed under ambient oxygen. It was concluded that the mechanism which controls nitrogen fixation rates in soybean nodules after exposure to an alteration in the external oxygen concentration involves the regulation of the nodule gas permeability. In Chapter VI the effects of soil temperature on nitrogen fixation rate and the nodule gas permeability are reported. It was found that diurnal cycles in nitrogen fixation were the result of changes in the soil temperature and were not related to daily fluctuations in photosynthetically active radiation. Changes in soil temperature also resulted in proportional changes in the nodule gas permeability such that a linear correlation between temperature and acetylene-reduction rates was maintained. These data are consistent with the hypothesis that nitrogen fixation is limited by the flux of oxygen entering the nodule, and in order for an increase in fixation rate to occur at elevated temperature, a proportional increase in nodule gas permeability must also occur. It was concluded that the change in gas permeability associated v;ith temperature could not be the result of a passive response due to changes in the diffusivity or solubility of oxygen in the nodule tissue, and was probably the result of an active mechanism which may regulate turgor pressure in the cortical cells. The data reported in these chapters indicate that the gas permeability of soybean nodules is a dynamic parameter which can respond to changing environmental conditions and which appears to be under the active physiological control. Data reported in Chapter VI for soybeans indicates that the primary regulatory mechanism for controlling nitrogen fixation rates on a diurnal basis may not be related to current rates of

PAGE 109

103 shoot photosynthesis, but does involve changes in the nodule gas permeability. These data on diurnal fixation rates are interesting in that they indicate that increased gas permeabilities result in higher fixation rates at least at elevated soil temperatures. Conversely, increased rates of nitrogen fixation which result from elevations in the rhizosphere oxygen concentration are seemingly not permitted in soybean as the nodule gas permeability adjusts to regulate these increased rates back to those originally observed under ambient conditions. Elevated nitrogen fixation rates are apparently maintainable in response to elevated temperature but not supra-ambient rhizosphere oxygen concentrations. Curiously, the nodule gas permeability increases in the former case to accommodate the elevated fixation rates, and decreases in the latter to prevent them. In Chapters V and VI it is speculated that control of the nodule gas permeability may be related to changes in turgor in the cells of the nodule cortex. A mechanism which would trigger turgor changes in response to the oxygen concentration or the temperature is currently unknown. This research points to the need for a more detailed understanding of the factors which control the gas permeability in nitrogen fixing root nodules. For example, if bacterial strains are developed for improved in -vitro nitrogen fixation, but the rate of oxygen diffusion into the infected nodules is determined by the nodule gas permeability, the increased potential for nitrogen fixation in these strains may not be realized in situ Attempts to improve symbiotic

PAGE 110

104 nitrogen fixation rates which do not effect the control of the nodule gas permeability may be ineffective. In conclusion, it has been demonstrated that the nodule gas permeability of fieldand hydroponically grown soybean is variable and responds to a number of environmental conditions. Furthermore, the magnitude of the gas permeability may be under physiological control as the responses reported here could not be explained in terms of a passive process. The nodule gas permeability appears to play an active and pivotal role in not only protecting nitrogenase from oxygen inactivation but also in controlling the rate of nitrogen fixation in symbiotic leguminous systems.

PAGE 111

APPENDIX BASIC PROGRAM FOR LAG-PHASE SIMULATION 100 REM ************************************************************* 200 REM 500 REM THIS IS A SIMULATION OF A LAG PHASE FOR ETHYLENE 550 REM PRODUCTION IN A SOYBEAN NODULE 600 J?EM 610 REM Assuming a solubility coef. for ^ ^2^4 ^ 660 REM 0.108, and a time constant of 2 se^. associated with 670 REM a 20 cm assay chamber with 600 cm /min flow rate. 675 REM At time 0, ^^^^ added to 10% to the gas supply, 678 REM perfect mixing in the chamber is assumed 680 REM 685 REM ************************************************************* 692 REM 693 REM ***************** OPEN FILES ******************************** 690 REM 695 REM OPEN THE OUT PUT FILE 700 FILE$ = "KM4.DAT" 850 OPEN FILE$ FOR OUTPUT AS FILE #1 900 REM 902 REM ***************** DECLARE THE VARIALBES ********************* 905 REM 1000 DECLARE DOUBLE CA(200), CE(200), Q(200), DCA(200) 1050 DECLARE DOUBLE DCE(200), EFLOW, DX, KONST, D, AREA, VN 1060 DECLARE DOUBLE R(5), CEAIR, DCEAIR 1100 REM 1120 REM •*>************** INITIALIZE ******************************** 1150 REM 1202 REM RADIUS OF BACTEROID VOLUME (CM) 1210 R(l) = 0.074 1212 R(2) = 0.092 1214 R(3) = 0.119 1230 REM 1245 SE = 0.108 1250 km = 0.004 1350 D = 4.6E-6 REM NODULE RADIUS OF 0.10 CM REM NODULE RADIUS OF 0.125 CM REM NODULE RADIUS OF 0.15 CM REM C^H, SOLUBILITY REM ^OR C2H2 REDUCTION REM DIFFUSIVITY OF C AT 1355 REM 26C IN CYTOPLASM 1360 REM 1398 REM **************** START ************************************* 1399 REM 1400 FOR RDX = 1 TO 3 : REM FOR 3 NODULE SIZES 1405 R = R(RDX) : REM BACTEROID ZONE RADIUS 1450 VN = 4*3.1416*R23/3 : REM BACT. ZONE VOLUME 1500 AREA = 4*3.1416*R-2 : REM BACT. ZONE SURFACE AREA 1548 REM 105

PAGE 112

106 1549 REM irllCJsjNtjbfa Ur Int iJAKKlijK 1550 FOR LDX = 1 TO 7 : REM rUK / i nlv-JsJNEib oto 1560 LENGTH = LDX .001 : n TT'yA KriM 1550 N — o ^ IjDX : NUMBER OF INCREMENTS 1 /UU Ua — liiMMUin/IM : riCji'i INCREMENTAL THICKNESS J/ DKJ \7r\J T — nV*ADT2i • VvJJjX Ua /\Kri/\ S rTlMPaPTWPWT VriT TTMP ^Ur'ilrMKliyiJ-jiN X V \JLjKJLrlCj louu JS-UlNo i U/ UA z : PPM PHMQTAMT FOR T.ATFP 1 Q tin uj. — u.uui : KrJ>i X XJYLH' XCMV^Ki-ji'liLiN X i r IJN — zuuu : KIMYI ZUUU ubAlK — u : KEiW. H J IN AIR IN CHAMBER INITIALIZE CONCENTRATIONS zU lU VM — ^.D.OEi fi^ VIM : Krjjyi ziuu r UK 1 — U i vj IN Z KEjyi r^T^ f T ^ — n • KTiM r p u 1 H_ J ( T \ — n • L,t ^ X ; — u : KiZjlYl r r u 1 i 2^4^ KCiK ^2 4 GENERATION KAlt, 0 inn KTPYT T O T O C\ Z J zo y ; 1 : REM DUMMY FOR 1ST ITTERATION Z JOU Eit JjUV. — u : REM C^H^ FLOW FROM CHAMBER Z4UU is. — U z4dU T = 0 : REM TIME IN SECONDS ,1 y1 C z44d REM zbUU K Cjti ^^^^^^^^^^^^^ AK i SOLUTION ************************** o cn c REM ZDDU irKXWi ffX ^ K JjUa XU S REM PRINT RADIUS, THICKNESS o c; on z-iyu zDy z KriM ZbUU K r.iA ^^^^^^^^^^^^^ ^ i AK i LOOP, PRINT DATA ******************* Rn RE^^ '^'7nn Z / UU PRINT ffl, T, CEAIR : REM PRINT TIME, CHAMBER [C^H^] ERAT = 10*CEAIR/Q ( N ) : REM C H FLOW/MAX FLOW TIME TO STOP ? ^ 2900 IF ERAT > =0.88 THEN GOTO 6000 : REM 3000 REM 3050 REM ************* LOOP, FIND SLOPE, PREDICT NEXT, REPEAT *** 3051 REM 3100 FOR TLOOP = 1 TO TFIN 3105 CA(0) = 0.10*(1-EXP(T/2)) : REM [C^H^] IN CHAMBER 3106 REM 3107 j^gy ********* pj^TES OF CHANGE *************************** 3110 REM T 1 on olzU REM CHANGE IN CONC. PER INCREMENT 3150 FOR I = 1 TO N-1 o o nn jzUU DCA(I) = K0NST*(CA(I-1) -2*CA(I)+CA(I+1)) Q(I)/VOLI TO cn ozoU DCE(I) = K0NST*(CE(I+1 ) -2*CE(I)+CE(I-1) ) + Q(I)/VOLI "3 "^nn NEXT I J Jul REM REM CHANGE IN CONC. C_H^ IN CHAMBER 3305 DCEAIR = 0.05*(EFLOW•10*CEAIR) 3310 REM PRODUCTION RATE 3350 Q(N) = VM*CA(N)/{CA(N)+km) 3360 REM CHANGE IN CONC. INSIDE BARRIER 3400 DCA(N) = {D*AREA/DX*( CACNl)-CA(N)) Q(N))/VN 3450 DCE(N) = (Q(N) D*AREA/DX *(CE(N)-CE(N-1) ) )/VN

PAGE 113

107 3451 REM 3452 REM ********* PREDICT NEW CONCENTRATIONS **************** 3453 REM 3455 CEAIR = CEAIR + DT*DCEAIR : REM [C^U^] IN AIR 3460 CE(0) = CEAIR*SE : REM [C K^] OUTER BARRIER EDGE 3470 REM 3475 REM [ ]'S IN INCRE14ENTS 3500 FOR I = 1 TO N 3550 CA(I) = CA(I) + DT*DCA(I) 3600 CE(I) = CE(I) + DT*DCE(I) 3650 NEXT I 3655 EFLOW = D*AREA/DX* (CE ( 1 )-CE (0 ) ) : REM C^H^ FLOW RATE 3700 K = K + 1 3750 T = DT* K : REM UPDATE TIME 3800 NEXT TLOOP 5000 GOTO 2600 : REM CHECK TIME, PRINT DATA 6000 PRINT #1, Q(N) : REM PRINT MAXIMUM RATE 6010 NEXT LDX : REM NEXT THICKNESS 6200 NEXT RDX : REM NEXT NODULE SIZE 7000 STOP : REM ALL DONE, GO HOME, PARTY! J i

PAGE 114

REFERENCES Altman, P. L., and D. S. Dittmer (eds.). 1971. Respiration and circulation. Federation of American Societies for Experimental Biology, Bethesda, Maryland. Appleby, C. A. 1984. Leghemoglobin and Rhizobixjm respiration. Ann. Rev. Plant Physiol. 35:443-478. Bergersen, F. J. 1962. The effects of partial pressure of oxygen upon respiration and nitrogen fixation by soybean root nodules. J. Gen. Microbiol. 29:113-125. Bergersen, F. J. 1971. Biochemistry of symbiotic nitrogen fixation in legumes. Ann. Rev. Plant. Physiol. 22:121-140. Bergersen, F. J. 1982. Root nodules of legumes: Structure and functions. Research Studies Press, New York, Bergersen, F. J., and D. J. Goodchild. 1973. Aeration pathways in soybean root nodules. Aust. J. Biol. Sci. 26:729-740. Bergersen, F. J., and G. L. Turner. 1975. Leghaemoglobin and the supply of 0^ to nitrogen-fixing root nodule bacteroids: presence of two oxidase systems and ATP production at low free concentrations. J. Gen. Microbiol. 92:345-354. Bergersen, F. J., and G. L. Turner. 1980. Properties cf terminal oxidase systems of bacteroids from root nodules of soybean and cowpea and of N^-fixing bacteria grown in continuous culture. J. Gen. Microbiol. 118:235-252. Berry, L. J., and W. E. Norris, Jr. 1949. Studies of onion root respiration II, The effect of temperature on the apparent diffusion coefficient in different segments of the root tip. Biochim. et Biophys. Act. 3:607-615. Carlson, T. 1911. The diffusion of oxygen in water. J. Amer Chem. Soc. 33:1027-1032. 108

PAGE 115

109 Criswell, J. G., U. D. Havelka, B. Quebedeaux, and R. W. F. Hardy. 1976. Adaptation of nitrogen fixation by intact soybean nodules to altered rhizosphere p02' Plant Physiol. 58:622-625. Criswell, J. G., U. D. Havelka, B. Quebedeaux, and R. W. F. Hardy. 1977. Effect of rhizosphere pO^ on nitrogen fixation by excised and intact nodulated soybean roots. Crop Sci 17:39-44. Davis, L. C. 1984. Diffusion of gases through plant tissues. Entry of acetylene into legume nodules. Plant Physiol. 76:854-857. Delieu, T., and D. W. Walker. 1981. Polaragraphic measurement of photosynthetic oxygen evolution by leaf discs. New Phytol. 89:165-178, Denison, R. F., and T. R. Sinclair. 1985. Diurnal and seasonal variation in dinitrogen fixation (acetylene reduction) rates by field-grown soybeans. Agron. J. 77:679-684. Denison, R. F., T. R. Sinclair, R. W. Zobel, M. M. Johnson, and G. M. Drake. 1983a. A non-destructive field assay for soybean nitrogen fixation by acetylene reduction. Plant and Soil. 70:173-182. Denison, R. F., P. R. Weisz, and T. R. Sinclair. 1983b. Analysis of acetylene reduction rates of soybean nodules at low acetylene concentrations. Plant Physiol. 73:648-651. Denison, R. F., P. R. Weisz, and T. R. Sinclair. In press. Oxygen supply to nodules as a limiting factor in symbiotic nitrogen fixation. In World Crops: Cool season food legumes. R. J. Summerfield (ed.). Martinus Nijhof, Dordrecht, Netherlands. Dungey, N. O. and N. J. Pinfield. 1980. The effect of tem.perature on the supply of oxygen to emJoryos of intact Acer pseudoplatanus L. seeds. J. Exp. Hot. 31:983-992. Eckhart, J. F., and C. A. Raguse. 1980. Effects of diurnal variation in light and temperature on the acetylene reduction activity (nitrogen fixation) of subterranean clover. Agron. J. 72:519-523. Fehr, W. R., C. E. Caviness, D. T. Burniood, and J. S. Pennington. 1971. Stage of development descriptions for soybeans. Glycine max (L.) Merrill. Crop Sci. 11:929-931. Goodchild, D. J, 1977. The ultrastructure of root nodules in relation to nitrogen fixation, p. 235-288. In International review of cytology. G, H. Bourne and J. F. Danielli (eds.). Academic Press, New York.

PAGE 116

110 Hardy, R. W. F., and U, D. Havelka. 1976, Photosynthate as a major factor limiting nitrogen fixation by field-grown legumes with emphasis on soybeans, p, 421-439, In Symbiotic nitrogen fixation in plants, P. S, Nutman (ed,), Cambridge University Press, New York, Imsande, J,, and E, J. Ralston. 1981. Hydroponic growth and the nondestructive assay for dinitrogen fixation. Plant Physiol. 68:1380-1384. Johnson, I. R., and J. H. M. Thornley. 1985. Temperature dependence of plant and crop processes. Ann, Bot, 55:1-24, Minchin, F, R,, J, E. Sheehy, and J, F. Witty, 1985, Factors limiting fixation by the legume Rhizobium symbiosis, p. 285-291, In Nitrogen fixation research progress. H. J. Evans, P. J. Bottomley, and W. E. Newton (eds.). Martinus Nijhoff Pviblishers, Boston. Minchin, F. R., J. F. Witty, J. E. Sheehy, and M. Muller. 1983. A major error in the acetylene reduction assay: Decreases in nodular nitrogenase activity under assay conditions. J. Exp. Bot. 34:641649, Orcutt, F, S,, and M, H, Seevers. 1937. A method for determining the solubility of gases in pure liquids or solutions by the Van SlykeNeill manometric apparatus. J. Biol. Chem. 117:501-507. Pankhurst, C. E., and J. I, Sprent. 1975a. Surface features of soybean root nodules. Protoplasma. 85:85-98. Pankhurst, C. E., and J. I. Sprent. 1975b, Effects of water stress on the respiratory and nitrogen-fixing activity of soybean root nodules, J, Exp, Bot, 26:287-304, Ralston, E, J,, and J, Imsande, 1982, Entry of oxygen and nitrogen into intact soybean nodules. J. Exp. Bot. 33:208-214. Robson, L., and J. R. Postgate. 1980. Oxygen and hydrogen in biological nitrogen fixation. Ann. Rev. Microbiol, 34:183-207. Schweitzer, L. E,, and J, E, Harper. 1980. Effect of light, dark, and temperature an root nodule activity (acetylene reduction) of soybeans. Plant Physiol. 65:51-56. Selker, J. M. L., and E. H. Newcomb. 1985. Spatial relationships between uninfected and infected cells in root nodules of soybean. Planta. 165:446-454. Sheehy, J. E., F. R. Minchin, and J. F. Witty, 1983, Biological control of the resistance to oxygen flux in nodules, Ann. Bot. 52:565-571,

PAGE 117

Ill Sinclair, R. S,, and J. Goudriaan. 1981. Physical and morphological constraints on transport in nodules. Plant Physiol. 67:143-145. Sinclair, T. R., and P. R. Weisz. 1985. Response to soil temperature of dinitrogen fixtion (acetylene reduction) rates by field-grown soybeans. Agron. J. 77:685-688, Sinclair, R. S., P. R. Weisz, and R. F. Denison. 1985. Oxygen limitation to nitrogen fixation in soybean nodules, p. 797-806. In World soybean research conference III: Proceedings. R. Shibles (ed.). Westview Press, Boulder. Sprent, J. I. 1972. The effects of water stress on nitrogen-fixating root nodules. II. Effects on the fine structure of detached soybean nodules. Mew Phystol 71:443-450. Tjepkema, J. D. 1971. Oxygen transport in the soybean nodule and the function of leghemoglobin Ph.D. Dissertation, University of Michigan, Ann Arbor. (Diss. Abstr. 32B:6248). Tjepkema, J. D., and L. S. Winship. 1980. Energy requirements for nitrogen fixation in actinorhizzal and legume root nodules. Science. 209:279-281. Tjepkema, J. D,, and C. S. Yocum. 1973. Respiration and oxygen transport in soybean nodules. Planta (Berl.). 115:59-72. Tjepkema, J. D., and C. S. Yocom. 1974. Measurement of oxygen partial pressure within soybean nodules by oxygen microelectrodes Planta (Berl.). 119:351-360. Weisz, P. R., R. F. Denison, and T. R. Sinclair. 1985. Response to drought stress of nitrogen fixation (acetylene reduction) rates by field-grown soybeans. Plant Physiol. 78:525-530. Witty, J. F., F. R. Hinchin, and J. E. Sheehy. 1983. Carbon costs of nitrogenase activity an legume root nodules determined using acetylene and oxygen. J. Exp. Bot. 34:951-963. Witty, J. F., F. R. Minchin, J. E. Sheehy, and M, I. Minguez. 1984. Acetylene-induced changes in oxygen diffusion resistance and nitrogenase activity of legume root nodules. Ann. Bot. 53:13-20. V/itty, J. F., L. Skot, and N. P. Revsbechs. 1985. Direct evidence for a variable barrier to diffusion into legume nodules. Abstr. 5-74. In Book Of Abstracts. 6th International Symposium On Nitrogen Fixation. Oregon State University, Corvallis. Winship, L. J., and J. D. Tjepkema. 1983. The role of diffusion in oxygen protection of nitrogenase in nodules of Alnus rubra. Can. J. Bot. 61:2930-2936.

PAGE 118

Winship, L. J,, and J. D. Tjepkema. 1985. Nitrogen fixation and respiration by root nodules of Alnus riibra Bong.: Effects of temperature and oxygen concentration. Plant and Soil. 87:91

PAGE 119

BIOGRAPHICAL SKETCH Paul Randall Weisz was born in Bryn-Mawr, Pennsylvania, in 1952. He graduated from Eckard College in 1970, with a Bachelor of Arts with honors and recieved a Master of Arts degree from Antioch, New England, in 1977. In 1983, he received his Master of Science degree from the University of Florida. 113

PAGE 120

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Dr. T. R. Sinclair, Chairman Professor of Agronomy I certify that I have read this study and that in roy opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Dr. S. L. Albrecht Assistant Professor of Agronomy I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Dr. T. E. Humphreys '' ^ Professor of Horticultural Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.

PAGE 121

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Assistant Professor of Horticultural Science This dissertation was sxabmitted to the Graduate Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 1986 iStllt Dean, ob/lege of Agriculture Dean, Graduate School