Arbuscular mycorrhizal colonization increases phosphorus uptake and growth of corn in a white clover living mulch system

Abstract

White clover (Trifolium repens L.) living mulch, a cover crop cultivation system, has been shown to improve phosphorus (P) nutrition and increase the yield and arbuscular mycorrhizal colonization rate of the main crop without inoculation of fungi. However, it remains unclear whether the P uptake of corn (Zea mays L.) in living mulch is promoted directly by arbuscular mycorrhizal colonization. We performed a pot experiment to test the hypothesis that living mulch increases the P uptake of corn by promoting arbuscular mycorrhizal colonization. The experimental design was a 2 × 2 factorial arrangement using fungicide treatment (fungicide application or no fungicide application) and cropping system (living mulch or no mulch). The fungicide dazomet was used to inhibit arbuscular mycorrhizal colonization. Without fungicide application, the P concentration and dry weight of the corn shoots were greater in the living mulch than in the no mulch treatment, indicating that living mulch improved the P nutrition and growth of corn. Fungicide application in living mulch, however, decreased the P concentration, dry weight of corn shoots, and arbuscular mycorrhizal colonization compared with no fungicide application. These results suggest that living mulch increases the P uptake of the main crop by promoting arbuscular mycorrhizal colonization by indigenous fungi.

Keywords:

• arbuscular mycorrhizae
• corn
• fungicide
• living mulch
• phosphorus.

1. Introduction

Arbuscular mycorrhizae (AM) increase the phosphorus (P) uptake and growth of host plants (Smith and Read 2008), and AM colonization is influenced by crop and soil management practices (Harrier and Watson 2003). A cover crop is a low-growing crop that provides soil protection and improves soil properties between periods of main crop production, and the cover crop is usually killed by using chemical herbicides before sowing the main crop (Fageria et al. 2005). Studies have shown that AM colonization of the main crop is facilitated by the introduction of a cover crop (Dodd and Jeffries 1986; Galvez et al. 1995; Boswell et al. 1998; Kabir and Koide 2000, 2002).

Living mulch (LM) is a cover cropping system in which a cover crop is planted either before or along with the main crop and maintained as a living ground cover throughout the growing season (Hartwig and Ammon 2002). In addition to having benefits similar to those of other cover cropping systems, white clover (Trifolium repens L.) LM can suppress weeds without the need for chemical herbicides (Uozumi et al. 2004). Reducing chemical inputs is one of the goals of sustainable crop production. Therefore, LM is useful in low-input, sustainable agriculture.

We previously reported that white clover LM increases the AM colonization rate of corn (Zea mays L.) and improves the P nutrition of corn during the early growth stage under field conditions (Deguchi et al. 2005, 2007). However, it remains unclear whether the improvement of P nutrition of corn in LM is promoted directly by AM colonization. According to Harrier and Watson (2003), AM colonization and the growth-enhancing effect of AM decreased with increasing soil P availability or with P fertilization. Thus, to understand how LM improves the P nutrition of the main crop and to develop plant nutrient management practices using LM, it is necessary to clarify whether AM colonization is the causal factor responsible for improving the P nutrition of the main crop. To clarify this point, researchers should compare crop growth in LM and in conventional cultivation with and without AM fungi. Under field conditions, it is difficult to apply fungicides to obtain non-mycorrhizal plants (Pedersen and Sylvia 1997). Therefore, in this study we performed a pot experiment in which we applied a fungicide to obtain non-mycorrhizal plants to test the hypothesis that AM colonization increases the P uptake of the main crop in an LM cropping system.

2. Materials and Methods

The pot experiment was conducted in 2009 in a greenhouse at Tohoku Agricultural Research Center, Morioka, Japan, using soils collected from an experimental field. The soil, classified as a Pachic Melanudand (USDA classification), was sampled from 0 to 10 cm depth, and its chemical properties were pH (H₂O) 5.8, total carbon 87.2 g kg⁻¹, total nitrogen (N) 6.3 g kg⁻¹, total P 1.8 g kg⁻¹, and available P 17.9 mg phosphorus pentoxide (P₂O₅) kg⁻¹ (Truog method). The experimental design was a 2 × 2 factorial arrangement using fungicide treatment (fungicide application or no fungicide application) and cropping system (LM or no mulch [NM]). Each treatment was replicated five times. To obtain non-mycorrhizal plants, we used the fungicide dazomet (98% 3,5-dimethyl-1,3,5-thiadiazinane-2 thione; Basamid®). To sterilize the soil, dazomet was applied at a rate of 400 g m⁻³ (6 g pot⁻¹) in December 2008, and the soil was stirred every 2 weeks for 2 months. The non-sterilized soil was also stirred at these times.

On 10 February 2009, cylindrical plastic pots (diameter 252 mm, depth 300 mm) were filled either with sterilized soil (11 kg fresh soil) for the fungicide-application treatments or with non-sterilized soil (11 kg fresh soil) for the no-fungicide-application treatments. On 13 February, P was applied to all pots as superphosphate at a rate of 200 kg P₂O₅ ha⁻¹ (1 g P₂O₅ pot⁻¹). On the same day, 1.0 g of white clover (T. repens cv. Huia) and 0.1 g of Rhizobium leguminosarum biovar trifolii, Mamezo® were mixed and sown in each LM pot (half the pots were fungicide-application pots and half were no-fungicide-application pots). Only 0.1 g of Mamezo® was sown in each NM pot. On 23 June, the white clover shoots in the LM pots were clipped and removed from the pots. Corn (Z. mays L. cv. 31N27) was then sown at a rate of six seeds per pot in all the pots. On the same day, N at a rate of 400 kg N ha⁻¹ (2 g N pot⁻¹) as ammonium sulfate and potassium (K) at a rate of 400 kg potassium oxide (K₂O ha⁻¹) (2 g K₂O pot⁻¹) as potassium chloride were applied to all pots to mask the effect of the LM on the N and K nutrition of corn. On 13 July, the number of corn plants per pot was thinned to four.

Surface soil samples (0–10 cm) were collected randomly from each pot before fertilization at sowing of the corn (23 June), and pH (H₂O) and available P content (Truog method) (Committee of Soil and Environment Analysis 1997) of the sieved soil samples (2-mm mesh) were analyzed. On 23 June, the dry weights of the white clover shoots were measured. At 29 days after sowing of the corn (DAS; 22 July), the dry weights of the corn shoots were measured. The P concentration and P uptake of the corn shoots were determined as described by Deguchi et al. (2007). White clover roots in a 7 × 7-cm sampling area were collected from each LM pot at sowing of the corn, and corn roots were collected from all pots at 29 DAS. The percentage of roots that were colonized by AM was determined as described by Deguchi et al. (2007).

A Student's t-test was used to compare the dry weight, P concentration, and P uptake of white clover shoots and AM colonization rate on white clover roots between the fungicide and no-fungicide treatments at sowing of the corn. Two-way analysis of variance (ANOVA) for a completely randomized design was used to examine the effects of the combinations of fungicide treatment (F) and cropping system (C) on the available P content of the soil at sowing of the corn, and the dry weight, P concentration, and P uptake of corn shoots and AM colonization rate on corn roots at 29 DAS. When the F × C interaction was significant (P < 0.05), means were compared using Tukey's test for multiple comparisons. Correlation analysis was used to establish the relationship between the P concentration and dry weight of corn shoots at 29 DAS.

3. Results and Discussion

In June 2009, white clover covered the soil surface of the LM pots. Upon removal from the pots (23 June), the dry weights of the white clover shoots in the fungicide and no-fungicide treatments were 62.0 and 49.4 g pot⁻¹ (P < 0.01), respectively. The P concentrations of the white clover shoots were 0.19% and 0.27% (P < 0.01), and P uptake by white clover shoots was 120.0 and 135.0 mg pot⁻¹ (P > 0.05) in the fungicide and no-fungicide treatments, respectively. After the shoots were clipped, the white clover regrew to cover the soil surface. The soil pH at sowing of the corn ranged from 5.7 to 5.9. Because the early growth of corn does not differ significantly across the pH range from 4.5 to 6.2 (Shimono 1990), soil pH likely did not affect the growth of corn in this study. The available P content of the soil at sowing of the corn is shown in Table 1. LM significantly decreased the available P content of the soil. In July, we observed a purple coloration of corn shoots, which is associated with P deficiency (Yamazaki 1966), except those plants in LM pots without fungicide application. The dry weight, P concentration, and P uptake of corn shoots at 29 DAS are shown in Table 1. The dry weight of corn shoots in LM pots without fungicide tended to be greater than that in NM pots with fungicide, and the value was significantly greater than that in NM pots without fungicide or in LM pots with fungicide. The P concentration and P uptake of corn shoots in LM pots without fungicide were significantly higher than those in the other treatments. The P concentration and dry weight of corn shoots were significantly correlated (r = 0.765, P < 0.001, n = 20). Together, these findings suggest that the low P acted as a limiting factor on the growth of corn in this soil, and that the corn plants, except those in LM pots without fungicide application, were P-deficient. Without fungicide application, the dry weight, P concentration, and P uptake of the corn shoots were all significantly higher in LM pots than in NM pots (Table 1). Thus, despite LM decreasing the available P content of the soil, P was supplied to corn in the LM treatments.

Table 1. Available phosphorus (P) content at sowing of the corn and the dry weight, P concentration, and P uptake of corn shoots and arbuscular mycorrhizal colonization rate on corn roots at 29 days after sowing

Treatment		Available P content (mg P2O5 kg^−1)	Dry weight (g plant^−1)	P concentration (%)	P uptake (mg plant^−1)	Colonization rate (%)
Fungicide application (F)	Cropping system (C)
Yes	NM	83.8 (13.1)	1.70 (0.12)^ab	0.21 (0.00)^c	3.57 (0.28)^b	1 (0.3)^c
Yes	LM	50.9 (5.2)	0.66 (0.07)^c	0.14 (0.01)^d	0.91 (0.07)^c	1 (0.2)^c
No	NM	78.6 (7.9)	1.57 (0.11)^b	0.27 (0.02)^b	4.19 (0.48)^b	42 (3.6)^b
No	LM	51.9 (8.8)	2.26 (0.25)^a	0.33 (0.01)^a	7.36 (0.80)^a	64 (1.9)^a
ANOVA
P value	F	0.712	<0.001	<0.001	<0.001	<0.001
	C	0.004	0.284	0.713	0.612	<0.001
	F × C	0.544	<0.001	<0.001	<0.001	<0.001
Values indicate the means (SE); n = 5.
P2O5, phosphorus pentoxide.
Different letters in each column indicate significant differences at P < 0.05 according to Tukey's test.
NM: no mulch; LM: white clover living mulch.

The AM colonization rates on white clover roots collected from the fungicide and no-fungicide pots on 23 June were 0% and 40% (P < 0.01), respectively. The fungicide application thus prevented AM colonization of the white clover roots. The AM colonization rate of the corn roots ranged from 1% to 64% at 29 DAS (Table 1). Thus, fungicide application also dramatically reduced the AM colonization of corn roots. In the no-fungicide pots, the AM colonization rate was significantly greater in LM than in NM. As noted above, in the no-fungicide pots LM significantly increased the P concentration and P uptake of the corn shoots compared with those in the NM treatments. Therefore, LM increased the P uptake of corn by promoting AM colonization, rather than by changing soil conditions. The white clover thus acted as a host for AM fungi at the time the corn was sown. In contrast, no AM hyphal networks were observed in the soil in NM pots when the corn was sown. Hyphae in senescent roots are important for the initiation of AM (McGee 1989). Therefore, without fungicide application, it appears that AM colonization occurred earlier in LM pots than in NM pots, with the result that the P uptake of corn shoots in LM was higher than that in NM.

In this study, we were able to obtain non-mycorrhizal corn in white clover LM by applying fungicide in a pot experiment. Our results clearly showed that AM colonization was essential for improving the P nutrition and growth of corn in white clover LM at 29 DAS. Previously, we reported that LM improved the P nutrition at 34 DAS and also increased the final yield of corn (Deguchi et al. 2007). We attributed the increase in the final yield to the promotion of AM colonization by LM at the early growth stage. The contribution of AM to plant P nutrition is reduced at high levels of available P (Thingstrup et al. 1998). Although LM also may not improve the P nutrition of the main crop when the soil P level is high, at medium and low soil P levels the use of LM would allow the rate of P fertilizer application to be reduced. Further studies are needed to evaluate LM as a fertilizer management technique by clarifying how much P can be supplied to the main crop by AM fungi in LM.

Acknowledgments

We thank Mr Yoichi Kodate for his technical assistance.