Plarlta (1984)162:316-326

P l a n t a 9 Springer-Verlag 1984

Nitrogen nutrition and the development and senescence of nodules on cowpea seedlings C.A. Atkins, B.J. Shelp, J. Kuo, M.B. Peoples and J.S. Pate Department of Botany, University of Western Australia, Nedlands, WA 6009, Australia

Abstract. Cowpea (Vigna unguiculata (L.) Walp cv. Vita 3) seedlings inoculated with Rhizobium strain CB756 were cultured with their root systems maintained in air or in Ar:O2 (80:20, v/v) during early nodule development (up to 24 d after sowing). Compared with those in air, seedlings in Ar:O 2 showed progressive N deficiency with inhibited shoot growth, reduced ribulose-l,5-bisphosphate carboxylase and total protein levels and loss of chlorophyll in the leaves. Nodule initiation, differentiation of infected and uninfected nodule tissues and the ultrastructure of bacteroid-containing cells were similar in the air and Ar: 0 2 treatments up to 16 d after sowing. Thereafter the Ar:O 2 treatment caused cessation of growth and development of nodules, reduced protein levels in bacteroids and nodule plant cells, and progressive degeneration of nodule ultrastructure leading to premature senescence of these organs. Provision of NO~ (0.1-0.2 mM) to Ar: O2-grown seedlings overcame the abovementioned consequences of N 2 deficiency on nodule and plant growth, but merely delayed the degenerative effects of Ar:O 2 treatment on nodule structure and senescence. Treatment of Ar:Oz-grown seedlings with NO~ greatly increased the protein level of nodules but the increase was largely restricted to the plant cell fraction as opposed to the bacteroids. By contrast, NO~- treatment of air-grown seedlings increased protein of bacteroid and host nodule fractions to the same relative extents when compared with air-grown plants not supplemented with NO~-. These findings, taken together with studies of the distribution of N in nodules of symbiotically effective plants grown from 15N-labeled seed, indicate that direct incorporation of fixation products by bacteroids A b b r e v i a t i o n ." RuBPCase

ase

= ribulose-1,5-bisphosphate carboxyl-

may be a critical feature in the establishment and continued growth of an effective symbiosis in the cowpea seedling. Key words: Nitrogen fixation - Nodule developm e n t - Senescence (nodules)- Vigna.

Introduction Development of the legume root nodule involves a complex series of structural and biochemical changes in host and invading Rhizobium, which finally results in a symbiosis in which the bacterial component provides reduced N and the host furnishes respirable substrates and a suitably regulated environment to support Nz-fixing activity by the bacteroids. Although infection-thread development and the accelerated mitotic activity attendant upon nodule tissue differentiation occur prior to the development of the functioning symbiosis, growth and maturation of nodule tissues usually continue long after the first appearance of nitrogenase activity (for a recent review, see Newcomb 1981). However, it is still not clear whether the provision of fixed N from the bacteroidal source has any direct bearing on the course of growth and development of the emergent nodule. Comparisons between ineffective and effective symbioses shed little light on the problem since in most ineffective associations nodule development ceases at or soon after the release of rhizobia from the infection thread (see review by Newcomb 1981) and before development of substantial nitrogenase activity. Although the rapid, premature senescence of ineffective nodules may be delayed in some cases by provision of exogenous combined N to the plant (MacKenzie and Jordan 1974), it remains uncertain whether it is the lack of nitrogenase activity

C.A. Atkins et al. : Nitrogen nutrition and nodule development in Vigna

in the bacteroid or some form of progressive incompatibility between host and Rhizobium which is critically damaging at this stage of nodule development. Rather than using ineffective mutants of Rhizobium which may be genetically altered in many characters in relation to an effective counterpart, we employ in this investigation an "inoperative" symbiosis which is generated by establishing a potentially effective symbiosis of cowpea under conditions in which the inoculated seedling root system develops in an atmosphere of Ar:O 2 (80:20, v/v) containing negligible N 2. The N nutrition and ultrastructure of the developing nodules on these Ar:O2-treated seedlings are then compared with the normal development of the same symbiotic association grown in air, extending the comparison to seedlings having access to N O r as well as to those growing solely on their cotyledonary stores of N. Materials and methods Plant material. Surface-sterilized cowpea seed (Vigna unguieulata (L.) Walp cv. Vita 3) (original source of seed, International Institute of Tropical Agriculture, Ibadan, Nigeria) was inoculated with a peat suspension of Rhizobium strain CB756 (Nitrogerm, Epping, NSW, Australia) and germinated in sand. Four d after sowing, groups of five seedlings were transplanted to 3.5-1 containers of N-free liquid culture solution (Layzell et al. 1979) maintained in a controlled-environment cabinet with a 12-h day at 30 ~ C and 800-1000 gmol m - 2 s 1 ( V H O 120 W coolwhite lamps Sylvania, Seneca Falls, N.Y., U S A and incandescent lamps) and a 12-h night at 20 ~ C. The lids of the culture containers were sealed at the edge and to the hypocotyl of each seedling with Terostat VII (Teroson G.m.b.H., Heidelberg, FRG), and a moisturized (80% relative humidity, 25 ~ C) stream of either CO2-free air or COz-free A r : O 2 (80:20, v/v) passed through the enclosed gas space above the nutrient liquid at a flow rate of 145-150 cm 3 min -1. Each day, degassed N-free nutrient solution was added to cultures through a small inlet port in such a way that there was negligible entry of air to the root atmosphere. The level of liquid was maintained just below the nodulated zone of the root system which, under these conditions of culture, is restricted to the main root and the first-formed three or four lateral roots (Layzell et al. 1979). In some experiments NO~ was supplied in the culture solution at 0.1 mM K N O 3 during the period 11-13 d after sowing, and then at 0.2 m M K N O 3 until the end of the study period. Pots of plants were removed from the cabinet at varying times from 11-24 d after sowing (i.e. from 7 to 20 d after transfer to liquid culture). Immediately upon opening, the culture containers the plants were harvested, separated into component organs, weighed, and used for various analyses.

15N-labeled plant material. Non-inoculated cowpea (cv. Caloona) plants were grown to maturity in sand culture with a nutrient solution (Atkins etal. 1980a) containing 1 0 m M K N O 3 enriched with 5 A % X (atom % excess) 15N. The harvested, mature seed, containing 4.68+_0.02 (standard error of mean; n = 10) A % X ~SN, was then inoculated with Rhizobium strain CB756, allowed to germinate, and the seedlings were

317

raised in N-free sand culture. Groups of 50 plants were harvested at 2- to 5-d intervals over the period from 14 to 28 d after sowing and the nodules collected for total-N and 15N analysis of their soluble and insoluble N fractions.

Gas analysis. The N 2 and O z contents of samples of the gas streams leaving the enclosed root atmospheres of the pots were measured using a gas-liquid chromatograph (GC-6AM, Shimadzu, Tokyo, Japan) equipped with a 2-m column of molecular sieve 5A (100-200 mesh; Waters Associates, Sydney, NSW, Australia) and a thermal conductivity detector. The CO2 levels in samples of the effluent gas streams were measured using an infra-red gas analyzer (Atkins and Pate 1977). Samples for analysis of gas composition were collected during a 1-h period in the middle of the light period of each day of the experiment. Extraction of nodules and assay of extracts. Freshly harvested nodules were held on ice prior to homogenisation in a chilled mortar and pestle with two volumes of 50 m M N-2-hydroxyethylpiperazine-Ni-2-ethanesulfonic acid (Hepes)-NaOH buffer (pH 7.5). The homogenate was passed through a a00-gm nylon mesh followed by centrifugation (10000 g, 20 min, 4 ~ C) and the soluble and particulate components collected as the supernatant and pellet, respectively. Assays of enzymes which serve as markers for bacteroids, plant-cell cytosol and plant-cell organelles in the extracts indicated that the supernatant consisted principally of plant-cell protein (see Atkins et al. 1980b). It is so termed in the text. The pellet fraction contained principally intact bacteroids, there being little evidence of bacteroid disruption under the conditions of extraction and centrifugation used (see Atkins et al. 1980b). The pellet fraction was then resuspended in homogenising buffer, passed twice through a chilled French pressure cell (Atkins 1974), and the protein, solubilized by disruption of the bacteroids, collected as the supernatant following centrifugation (10000 g, 20 rain, 4 ~ C). This fraction is referred to as "bacteroid protein" in the text. Protein was measured by the method of Lowry et al. (1951), using bovine serum albumin (Commonwealth Serum Laboratories, Melbourne, Vic., Australia) as standard. Nodules from plants grown from 15N-enriched seed were extracted as described above except that the bacteroid-containing pellet fraction was not treated with a French press. The supernatant and pellet fractions (termed "soluble" and "insoluble", respectively) were assayed for total N by a standard Kjeldahl procedure and for 1sN by optical emission spectrometry (N-150 spectrometer; Jasco, Tokyo, Japan) following Dumas combustion in quartz capillaries (Perschke et al. 1971).

Analysis of leaf tissue. Measurements of the soluble protein, chlorophyll a + b and ribulose-l,5-bisphosphate carboxylase (RuBPCase) protein contents of the unifoliolate pair of leaves and the leaflets of the first trifoliolate leaf were carried out as described in Peoples et al. (1983).

Light and electron microscopy. For light microscopy, whole or halved nodules were fixed (Karnovsky 1965) at pH 6.9 for 12-24 h at room temperature. The material was subsequently dehydrated in a graded ethanol series, embedded in glycol methacrylate (O'Brien and McCully 1981), and 2.5-gm-thick sections were stained with toluidine blue 0 (0.05%, w/v, in pH 4.4 benzoate buffer) and amido black 10 B (1%, w/v, in 7 %, v/v, acetic acid) to stain for protein, or by periodic acid-Schiff's (PAS) reaction to visualize starch grains and mucopolysaccharides. The PAS-stained sections were counterstained with either toluidine blue 0 or amido black a0B. Stains and embedding materials were obtained from BDH Chemicals, Kew East, Vic., Australia.

C.A. Atkins et al. : Nitrogen nutrition and nodule development in Vigna

318

Results

For transmission electron microscopy small segments of nodules were fixed as above, post-fixed in 1% (w/v) OsO,~ in 25 m M potassium-phosphate buffer (pH 6.9), dehydrated in graded ethanol, and embedded in epoxy resin (Spurr 1969). Thin (1 btm) sections were stained with uranyl acetate and lead citrate.

Gaseous composition of atmospheres surrounding nodulated root systems. Provision of a flowing gas stream containing 80% Ar and 20% O 2 to the sealed liquid-culture containers resulted in very low levels of N 2 (0.2-0.5%, v/v) in the atmosphere surrounding the nodulated roots (see Table 1). The rooting space of seedlings treated with A r : O 2 contained levels of O 2 comparable to those of seedlings treated with a stream of air (Table 1). The CO2 concentration of gas leaving the pots varied markedly on a diurnal basis and increased with age of the plants. The range of values recorded throughout experiments was from 300 to 700 gl

Table l. N 2 and 0 2 contents of the effluent gas stream from

the sealed root atmospheres of nodulated cowpea plants receiving either an air or an A r : O 2 stream Gas mixture passed through root environment

Concentration in effluent gas stream (%, v/v)

Air 80 A r : 2 0 O 2

N2

02

73.7 -78.7 ~ 77.33 -+ 0.20 b 0.23-0.51 0.375_+0.012

19.6 -21.8 20.87-t:_ 0.11 18.4 -20.8 19.79___0.11

C021-1

Growth and development of plant shoot organs. Measurements of fresh weight, total chlorophyll, soluble protein and RuBPCase protein were made for the unifoliolate and the first trifoliolate leaves of the seedlings of each treatment (Fig. 1). These organs constituted the bulk of the photosyntheti-

Range of values in effluent gas stream from pots in three separate experiments sampled on 10 occasions from 6-23 d after sowing b Mean_+ standard error of the mean ( N = 30)

i

i

i

i

~ .57 ~ ~

/i10 /

I. /

///

/

I

I,... e-

-fi4

e~

/

iI i I 01

,?./o-

O

E v

o

,2>"

4m

,

/

to

,o

I,~ eCtl

,,,,,;o ,>~..-"O ~ o

Q

i

B

A

9

o---o 9

l

I

7.i'm ~.

Ar:O2+NO 3

I

I

I

18

i

i

/I

p2--o

/i//I t"

E

I

i

e-

L

I

I

18

I

.o

I

22

I

I

I

O. O~2C

,-:~

E

o //o~/" I

O.

> X G

/.e; \

-~ 2o .Q

I .~

o~.

A Ill

O u}

ee

S-I

14

I

I

18

L

I

I

22

days

14 after sowing

I

D

r162

-.,

i

I

14

1

P\.

40

I

22

i

i

O'------'.O~o~ e

O

9 Ar:O 2

C

A

ee~

a i r + NO 3

-,

9 --- 9

14

>2

air

o~o

eva

-

I

I

18

-

O ~

p

o

I

22

Fig. l A D . Effect of air or A r : O 2 (80:20; v/v) around the nodulated root systems of water-cultured cowpea on A fresh weight of shoot; B chlorophyll a + b of leaves; C soluble protein of leaves; and D RuBPCase protein of leaves. The air or A r : O 2 gas streams were maintained from 4 to 24 d after sowing and NO 3 was added, where indicated, to give a final concentration in the culture solution of 0 . 1 m M from l l t o l 3 d a n d 0.2 m M from 14 to 24 d

C.A. Atkins et al. : Nitrogen nutrition and nodule development in

Vigna 20

e~

0.2

o--o

air

o--o

air+NO 3

o--e o--o

Ar:O 2 Ar:O2+NO 3

..t~~ /0~=.__

......

~ 0 il

319

[]

AIR

bacteroid

[~

plant cell

O3

O z

a

-F

133

._~

y:

10

u~

0.3

,r= I

E

17

14

19

21

24

21

24

20 l A r : O 2 a..,..lt-~ll

10

I

I

14

1

__

I

18

I

I

I

22

days after sowing

cO

O

root system of water-cultured cowpea on fresh weight of nodules. Treatments were as outlined in Fig. 1. Bars indicate average SE of means up to 18 d (1) and from 18 to 24 d (2)

e~

Growth and development of nodules. Although microscopic examination of roots showed visibly

I

O3

Fig. 2. Effect of air or Ar: 0 2 (80 : 20; v/v) around the nodulated

cally active surface of the plants throughout most of the experiment, the second trifoliolate leaf emerging between 20 and 22 d after sowing. Treatment with Ar:O 2 resulted in a severe inhibition of increase in leaf fresh weight (Fig. 1 A) and a marked reduction in production of chlorophyll (Fig. 1 B) after 16 d. Increases in total leaf soluble protein (Fig. 1 C) and RuBPCase protein (Fig. 1 D) were also reduced after 16 d. The severe N deficiency imposed on shoot development by Ar:O2 treatment was readily overcome with a low level of NO~ in the nutrient solution (Fig. 1; Ar:O2 + NO~ treatment). Provision of NO~ to air-grown seedlings (air + N O ~ treatment) increased each of the above components in comparison with the air treatment without NO~, indicating that shoot development of plants relying solely on symbiotically fixed N was probably limited by N supply. The "nitrogen-hunger" state of both air- and Ar:O 2grown seedlings was evident from the net losses of protein from leaves over the period 18-24 d. The effect of added NO~ was especially pronounced on levels of leaf protein and RuBPCase both of which were doubled compared with airgrown plants (Fig. 1 C, D). Except for the level of chlorophyll, which was stimulated by added NO~ to a greater extent in air-grown than in Ar: O2-grown plants (Fig. I B), the other measured components of leaf development were very similar in the A r : O 2 + N O ~ and a i r + N O ~ plants (Fig. 1 A, C, D).

i

e-

O z

+

14

17

19

days after sowing

Fig. 3. Effect of air or Ar: O2 (80 : 20; v/v) around the nodulated root systems of water-cultured cowpea on soluble protein content of the plant cells and bacteroids of nodules. Treatments were as outlined in Fig, 1

emergent nodules as early as 8 d, nodule tissue could not be sampled effectively until 12 d after sowing. Up to 14 d, the air and Ar:O 2 treatments with or without added NO~ showed similar rates of fresh-weight increase of the nodules, but after this time those in Ar:O2 alone grew more slowly, achieving a fresh weight at 24 d of about 75% of that of the other treatments (Fig. 2). The reduction in nodule mass caused by Ar:O 2 was overcome by NO~ addition but, unlike the case with shoot organs (Fig. 1), growth of nodules of NO~-treated plants was not stimulated above that of seedlings grown in air without NO~ supplementation. Treatment with Ar:O z caused significant reduction in the protein content of the nodule (Fig. 3). This was obvious at 14 d and became more pronounced thereafter. Although both bacteroid-located and plant proteins were reduced by Ar:O 2 treatment, the decline was more marked in the plant fraction. Bacteroid protein consequently constituted the major component (up to 70%) of the total protein of nodules cultured in Ar:O2, compared with around 40% in those grown in air (Fig. 3). Addition of NO 3 to culture

C.A. Atkins et al. : Nitrogen nutrition and nodule development in Vigna

320 I

f

I

i

F

]

f

I

o

~

//< "

2

9

ID

--

/// .~;~/" insoluble

0.4 3 Z

?

~

i

O.6

. -.2 2-

=

_=

o E

insolul~le\ ~

.~g"

"~ 3"

e_

I

14

I

I

o_

I

i

I

18 22 days after sowing

I

I

I

26

Fig. 4. Distribution of ~SN and total N between the soluble

(principally plant cell) and insoluble (principally bacteroid) fractions of nodules of cowpea seedlingsgrown from 15N-enriched seed (4.68A%X lSN) in sand culture solutions caused a progressive increase of total nodule protein in air-grown plants, resulting, at 24 d, in nodules containing around 30% more protein in the air + NO~ treatment than in air alone. In Ar:O2-grown plants the increase caused by NO~- was proportionately greater, with maximal total nodule-protein levels being more than five times those in nodules from Ar: 0 2 plants not supplemented with NO~-. Addition of NO~- to Ar: 0 2 plants resulted in increases in both the bacteroidand plant-protein fractions of nodules, but the increase in the plant-protein fraction was proportionately much greater, ultimately reaching a maximum level (at 19 d) up to ten times that of untreated Ar: O2-grown plants and more than double that of nodules of normal symbiotic plants (Fig. 3). Despite the initial stimulation of nodule-protein synthesis by NO~- in plants grown with Ar:O2, the protein levels declined after 19 d. This decline was not a feature of nodules of the a i r - or air + NO~--grown plants.

Distribution of lSN in nodules of plants cultured from 15N_enriched seed. Both the "soluble" (primarily plant cytoplasm) and "insoluble" (principally bacteroid) fractions of nodules from ~SN-labeled plants showed considerable enrichment by labeled N at the earliest sampling time (14 d after sowing), as would be expected from the nodule's early dependence on cotyledonary N. Thereafter, rapid dilution of both N fractions of the nodule occurred as non-enriched, fixed N contributed to

rapidly increasing pools of N in the nodule (Fig. 4). Throughout the period 14--24 d, lSN enrichment of the "soluble" plant fraction of the nodule indicated that there was a greater proportional contribution of cotyledon N to protein synthesis of plant cells and a higher proportional investment of newly fixed N in bacteroids.

Ultrastructure of nodules. Up to about 16 d, the structure of nodules cultured in air was indistinguishable from those grown in Ar:O 2 (Fig. 5A, B). In each case uninfected cells contained large starch grains while the grains in infected cells were relatively small. Later in development differences became apparent between the treatments, starch grains remaining in the bacteroid-containing cells of air-grown nodules but not in those of Ar:O 2grown nodules (Fig. 5 C, D), and amido-black-positive protein being at a much higher level in the bacteroid-containing cells of air-grown (Fig. 5 E) than Ar: O2-grown nodules (Fig. 5 F). Electron-microscopic examination of the ultrastructure of infected nodule cells likewise showed little difference between treatments up to 16 d after sowing (Fig. 6A, B). Thus, apparently unaffected by Ar:O z treatment up to this stage were the extent of cell vacuolation, the size and distribution of bacteroids, formation of peribacteroid membranes, the frequency of mitochondria and plastids, and the association of the latter with large starch grains in uninfected cells. Later in development, however, there was a progressive degeneration of the infected cells in nodules cultured in Ar:O 2 (Fig. 7). The cytoplasm became less dense and plant cell organelles disappeared, until, at 24 d, little or no cytoplasm remained. At this time, ribosomes, plastids, mitochondria and endoplasmic reticulum, which were all readily discernible in the cells of air-grown nodules (Fig. 7A, C), could no longer be seen in those grown in Ar:O 2 (Fig. 7B, D). Furthermore, peribacteroid membranes of Ar: O zgrown nodule cells were not obvious, bacteroid ribosomes had disappeared, and bacteroid chromatin material had become more concentrated and electron-dense (Fig. 7 D). Addition of NO~ to Ar: Oz-grown cultures arrested, to some extent, degeneration of the plant cells. Even after 24 d the cytoplasm was retained and the peribacteroid membranes remained intact (Fig. 8). There was however a reduction in the number of intact organelles in the plant cells. Consistent with the disappearance of small starch grains in periodic-Schiff-stained material (Fig. 5 D) plastids were not found in the infected cells. Compared with Ar:O2-grown nodules (Fig. 7D), bac-

R

.r

i n

~ e#

200~m,

f

a ~4 e

4~

Fig. 5 A-E. Light micrographs of thin sections of nodules from cowpea plants cultured with their nodulated root systems maintained in air or in 80% Ar:20% 0 2 (v/v). A, B Typical of nodules from plants 16 d after sowing and cultured in either air or Ar:O 2. Stained for starch with PAS reagent. Infected, bacteroid-containing cells (B) showed small starch grains (arrowed) while uninfected cells (U) showed large grains. In At: 02 and air-grown nodules, the distribution of central and cortical tissues, containing peripheral vascular bundles (V) was similar. C Nodule from plants 24 d after sowing and cultured in air. Distribution of starch grains was similar to that of 14-d nodules, with infected cells (B) containing small starch grains (arrows). D Nodule from plants 24 d after sowing and cultured in Ar:O 2. Distribution of starch grains in uninfected cells was similar to that in air but small starch grains in infected cells (B) were absent. E Nodule from plants 24 d after sowing and cultured in air. Section stained for protein by amido black. F As for E but plants cultured in Ar:O2. Magnifications for A, C, D = x 500; for B, E, F = x ~25

Fig. 6A, B. Electron micrographs of thin sections of nodules from cowpea plants 16 d after sowing and cultured with root systems maintained in A air; or B A r : 0 2 (80 : 20; v/v). S = small starch grain of infected cells; S (circled) = larger starch grain of uninfected cells; V=vacuole; N=nucleus; B=bacteroid; U = uninfected cell. x7500

Fig. 7 A-D. Electron micrographs of thin sections of nodules from cowpea plants 24 d after sowing and cultured in A, C, air; or B, D, Ar: O 2 (80 : 20; v/v). B = bacteroid; P = plastid; M = mitochondrion; W = cell wall; E = endoptasmic reticulum. Arrowheads indicate perihacteroid membrane and asterisks the lack of cytoplasm in Ar:O2-grown nodules. Magnification for A, B = x 9000; for C, D = x 36000

324

C.A. Atkins et al. : Nitrogen nutrition and nodule development in Vigna

Fig. 8. Electron micrograph of thin section of a nodule from cowpea plants 2 4 d after sowing, cultured in Ar:O~ (80:20; v/v) and supplemented with NO~ added to the culture solution. Arrowheads indicate peribacteroid membrane. B = bacteroids; S (circled) = starch; E = endoplasmic reticulum ; M = mitochondria. • 36 000

teroids in nodules supplemented with NO~- showed little change in fine structure (Fig. 8) although ribosomes were not as obvious as in those from airgrown nodules (Fig. 7C). Discussion

Nitrogen nutrition of developing nodules. Although continuous A r : O 2 treatment of the root environment of nodulated cowpea did not entirely eliminate N 2 from the rooting atmosphere, the highest concentrations recorded (0.51%; Table i) were very low compared with the in-vivo KIn(N2) of nitrogenase (10%; Pate et al. 1984) and, while possibly still of some morphogenetic relevence, were clearly unimportant in terms of the nutritional requirement for this element in plants relying solely on N 2 fixation. Cultivation of nodulated cowpea in A r : O 2 and in the absence of any other exogenous N source accordingly resulted in progressive

N deficiency with markedly inhibited shoot growth, reduced leaf-protein levels and chlorophyll breakdown. Under these conditions growth and development of nodules was arrested. Protein tevels in both plant cells and bacteroids were markedly reduced and changes in nodule ultrastructure were consistent with the premature senescence of these organs. Although the consequences of N z deficiency on shoot growth and nodule fresh weight were overcome by the addition of NO 3 to the root system, the " e x t r a " N supply did not entirely eliminate the degenerative effects on nodule structure and merely delayed, for a short time, the onset of premature senescence. Addition of NO~ enhanced the supply of N to nodules (Fig. 3) whether the root atmosphere supported a high or a very low rate of N 2 fixation. In air-grown nodules both bacteroid and plant-cell protein levels were increased to the same relative extent by the addition of NO~-, but in the A r : O 2-

C.A. Atkins et al. : Nitrogen nutrition and nodule development in

grown, non-fixing nodules, addition of NO3 caused marked increase of the plant-cell protein but not of the bacteroid protein. A simple interpretation of the disproportional distribution of protein resulting from NO~ addition to Ar: O2-treated nodules is that the extra N entering the nodule from the host following assimilation of NO 3 was readily available for plant-cell protein synthesis but did not reach the bacteroids or was not readily utilized in their metabolism. It is therefore suggested that ~ self-feeding" of bacteroids, using ammonia currently produced by nitrogenase, may be a critical feature of the establishment and continued growth of an effective symbiosis in cowpea. Support for this concept was obtained in the study of 15N dilution in cowpea seedlings grown from 15N_enriched seed. An appreciably greater proportional dilution of 15N in the bacteroids than in the surrounding host plant cells was evident over the period 14-21 d, indicating selective incorporation of newly fixed N within bacteroids and preferential utilization of cotyledonary N rather than fixed N into the plant component of the nodule (Fig. 4). The patterns of protein distribution in nodules cultured in Ar:O 2 were, however, likely to have been complicated by premature senescence of the organ. As N deficiency became severe, differential changes could have occurred in the functioning of protein-synthesising systems in the plant and bacteria or in the permeability of the peribacteroid or bacterial cell membranes. Therefore Ar:O ztreated nodules may well have experienced a more severe restriction of N movement to their bacteroids than in those grown in air, as indeed is evident when comparing the distribution of 15N and total N in normal, air-grown nodules of the 1SN-labeled seedlings with that of the N partitioning within nodules of the Ar:Oz-treated plants. Further testing of this hypothesis will require critical 15N-labeling experiments to establish more clearly the proportional contribution of cotyledon N, fixed N and the distribution of added exogenous sources of N to plant-cell and bacteroid development during establishment of the symbiosis.

Nitrogen deficiency and senescence of nodules." Nodule degeneration in legumes may occur naturally as a result of changes in the environmental factors affecting plant growth or changes in plant growth habit (see review by Sutton 1983). Although the details and timing of ultrastructural and biochemical modification of nodules have not been clearly established for many of the examples of premature senescence which have been described, the general

Vigna

325

ultrastructural pattern of change is one in which the plant cytoplasm becomes relatively devoid of organelles followed by degeneration of the peribacteroid membranes, and finally changes in bacteroid fine structure (Vance et al. 1980; Tu 1975; Newcomb 1981; Sutton 1983). In essence this pattern was observed here following Ar:O2 treatment. Similar changes have been observed in the degeneration of ineffective nodules (MacKenzie and Jordan 1974; Werner et al. 1980) where severe N deficiency, like that caused by Ar:O2, might be inferred. The Ar:O 2 treatment differed however from many ineffective symbioses (MacKenzie and Jordan 1974; Werner et al. 1980) in that starch accumulation in nodule cells was no greater than normal. Presumably the rates of respiration supporting the biochemical activities of nodules in Ar:O2 were sufficient to prevent excessive accumulation of translocated sugar. For instance, nitrogenase activity supporting H; evolution still takes place in the Ar:Oz-treated nodule (Pate et al. 1984). Basset et al. (1977a, b) demonstrated that, to some extent, severe degeneration of nodule plant cells of an ineffective soybean symbiosis could be retarded by addition of minimal levels of N to the plant culture and suggested that the plant response was one of N starvation and not specifically attributable to the ineffective strain of Rhizobium. A similar response was shown in the ultrastructure of infected cells of Ar:O2-grown nodules following additions of NO~ to the rooting medium which were sufficient to relieve N deficiency in the plant shoot as well as in the nodules. Despite the sufficiency of N and carbohydrate in these nodules, senescence was not prevented, a progressive loss of both bacteroid and plant cell protein occurring after 19 d (Fig. 3). These observations are not in accord with the generally held concepts of the plant's "carbohydrate supply" or "' carbohydrate-nitrogen relationship" as being essential determinants of nodule senescence (for a review, see Sutton 1983). Similarly, a more recent proposal by Sutton (1983), that senescence is caused by ammonia toxicity under conditions of carbohydrate limitation, would seem inappropriate to explain all responses described in this paper. The possible role for plant growth regulators, formed by either partner to the symbiosis, in promoting or retarding nodule senescence cannot be overlooked; however, the response of nodule development to Ar: O 2 treatment indicates an all-important role for ammonia production by the N2-fixing microsymbiont in the progressive enlargement and maintenance of an effective symbio-

326

C.A. Atkins et al. : Nitrogen nutrition and nodule development in Vigna

sis. In addition to the direct incorporation of fixed N by bacteroids, excretion of ammonia may affect the permeability of the bacteroid and peribacteroid membranes, and thus the exchange of other solutes between the plant cell and the bacteroid. Growth of nodulated plants under conditions of N 2 deficiency would appear to offer a useful means of generating a readily manipulated "ineffective" symbiosis in which it becomes possible to distinguish more clearly the separate roles of bacterium and host in nodule development and senescence. We would like to thank Paul Storer and Paul Sanford for skilled technical assistance. The work was supported by the Australian Research Grants Scheme (to C.A.A.), the Wheat Industry Research Council of Australia (to C.A.A. and J.S.P.) and by a National Science and Engineering Research Council of Canada postdoctoral fellowship (to B.J.S.).

References Atkins, C.A. (1974) Occurrence and some properties of carbonic anhydrases from legume root nodules. Phytochemistry 13, 93-98 Atkins, C.A., Pate, J.S. (1977) An IRGA technique to measure CO z content of small volumes of gas from the internal atmospheres of plant organs. Photosynthetica 11,214-216 Atkins, C.A., Pate, J.S., Griffiths, G.J., White, S.T. (1980a) Economy of carbon and nitrogen in nodulated and nonnodulated (NO3-grown) cowpea (Vigna unguiculata (L.) Walp). Plant Physiol. 66, 978-983 Atkins, C.A., Rainbird, R.M., Pate, J.S. (1980b) Evidence for a purine pathway of ureide synthesis in Nz-fixing nodules of cowpea (Vigna uriguiculata (L.) Walp). Z. Pflanzenphysiol. 97, 249-260 Bassett, B., Goodman, R.N., Novacky, A. (1977a) Ultrastructure of soybean nodules. I. Release of rhizobia from the infection thread. Can. J. Microbiol. 23, 573-582 Bassett, B., Goodman, R.N., Novacky, A. (1977b) Ultrastructure of soybean nodules. II. Deterioration of the symbiosis in ineffective nodules. Can. J. Microbiol. 23, 873-883 Karnovsky, M.J. (1965) A formaldehyde-glutaraldehyde fixa-

tive of high osmolarity for use in electron microscopy. J. Cell Biol. 27, 137A Layzell, D.B., Rainbird, R.M., Atkins, C.A., Pate, J.S. (1979) Economy of photosynthate use in nitrogen-fixing legume nodules. Observations on two contrasting symbioses. Plant Physiol. 64, 888 891 Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275 MacKenzie, C.R., Jordan, D.C. (1974) Ultrastructure of root nodules formed by ineffective strains of Rhizobium meliloti. Can. J. Microbiol. 20, 755-758 Newcomb, W. (1981) Nodule morphogenesis and differentiation. Int. Rev. Cytol., Suppl. 13, 247-298 O'Brien, T.P., McCully, M.E. (1981) The study of plant structure. Principles and selected methods. Termarcarphi Pty., Melbourne, Australia Pate, J.S., Atkins, C.A., Layzell, D.B., Shelp, B.J. (1984) Effects of N 2 deficiency on transport and partitioning of C and N in a nodulated legume. Plant Physiol. (in press) Peoples, M.B., Pate, J.S., Atkins, C.A. (1983) Mobilization of nitrogen in fruiting plants of a cultivar of cowpea. J. Exp. Bot. 34, 563 578 Perschke, H., Keroe, E.A., Proksch, G., Muehl, A. (1971) Improvements in the determination of nitrogen-15 in the low concentration range by emission spectroscopy. Anal. Chim. Acta 53, 459 463 Spurr, A.R. (1969) A low-viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res. 26, 31-43 Sutton, W.D. (1983) Nodule development and senescence. In: Nitrogen fixation, vol. 3: Legumes, pp. 144-212, Broughton, W.J., ed. Clarendon Press, Oxford, UK Tu, J.C. (1975) Rhizobial root nodules of soybean as revealed by scanning and transmission electron microscopy. Phytopathology 65, 447-454 Vance, C.P., Johnson, L.E.B., Halvorsen, A.M., Heichel, G.H., Barnes, D.K. (1980) Histological and ultrastructural observations of Medicago sativa root nodule senescence after foliage removal. Can. J. Bot. 58, 295-309 Werner, D., Morschel, E., Stripf, R., Winchenbach, B. (1980) Development of nodules of Glycine max infected with an ineffective strain of Rhizobium japonicum. Planta 147, 320-329

Received 2 March; accepted 7 May 1984