Abdessamad Fakhech·Lahcen Ouahmane·Mohamed Hafidi,2

Abstract Atlantic forest natural soil flora was tested on two leguminous pioneer species Acacia gummifera and Retama monosperma to be used in rehabilitation programs of the coastal sand dunes forest of the Essaouira region.The rhizospheric soil of two endemic plant species:R.monosperma and Juniperus phoenicea was sampled and split into two categories,one from native rhizospheric soil,the other of the sterilized rhizospheric soil.Investigation was focused on mycorrhizal formations,but other forms of beneficial symbiosis such as rhizobia and viable soil microflora were also studied.Growth and nutrition variables assessed included lengths of roots and shoots,number of branches,ratio of root to shoot dry mass and water,nitrogen and phosphorus levels.Results showed important mycorrhizal associations in roots of both plants,presence of nodules and abundance of viable soil microflora.J.phoenicea had a 100% frequency of mycorrhizal formation and an intensity of 80% compared to a frequency of 80% and intensity of 54% for R.monosperma.Nodules had the same density of CFU regardless of the origin.Abundance of viable microflora in rhisospheric soil of R.monosperma,J.phoenicea and the control differed significantly.Among the studied variables for A.gummifera and R.monosperma,nitrogen and phosphorus uptakes significantly wit the use of the nonsterilized rhizospheric soil.Both species doubled their phosphorus uptake when colonized by mycorrhizal species,R.monosperma doubled its nitrogen uptake and A.gummifera increased it by seven times compared with the control.No significant difference was noted for the other variables.

Keywords Nitrogen·Phosphorus·Juniperus phoenicea·Acacia gummifera·Retama monosperma·Mycorrhizae·Rhizobia

Introduction

Degraded forests are commonly restored using pioneer tree or shrub plantations.Promoting success and development of such plantations could be achieved using microorganisms present in forests soils.Forest soils are characterized by rich microbial communities and nutrient availability.They are a simple tool to enhance growth of plants in afforestation plans.The studies addressing plant growth response to natural soil inoculum are scarce(Johnson 1993;Kiers et al.2000).Forests soils are usually used to restore exhausted and abandoned agricultural areas where they show great potential compared to agricultural soils(Richter and Stutz 2002;Williams et al.2011;St-Denis et al.2017),but they are rarely trialed in forest restoration.

When considering stress-resistant,competitive plantations in rehabilitation programs,native species are usually prioritized,but non-native species can play an important role in some cases(Jacobs et al.2015).Pioneer species perform better than other successional groups in forest restoration(Zangaro et al.2000).

Although coastal sand dune ecosystems can have considerable vegetation cover,mycorrhizae are necessary for the ecosystem existence(Koske and Poison 2011).They contribute to the acquisition of essential nutrients such as nitrogen(Duponnois et al.2005;Jin et al.2005)and phosphorus(Schachtman et al.1998)and increase plant adaptation to the harsh conditions of the coastal sand dunes by contributing to the ecosystem biodiversity(Van Der Heijden et al.1998).Arbuscular mycorrhizae(AM)help plants become established,resist pathogens and tolerate various stresses in disturbed sand dunes(Nicolson 1960;Koske et al.2004;Koske and Poison 2011).

Performance improvement of dunes plant species seems to be very helpful in improving the success rates of the dunes rehabilitation programs.One way to do that is by exploiting the symbiotic association between roots of plants and some microorganisms such as mycorrhizal fungi and plant growth promoting rhizobacteria(PGPR)in the soil.

Mycorrhizae,a formation between roots of plants and a specialized fungi,are a key to this enhanced performance(Pasqualini et al.2007).The selected plants for the present study are Acacia gummifera(Moroccan gum tree)and Retama monosperma(retam).Retam is a shrub that very quickly produces an important,effective ground cover.Acacia gummifera was chosen to be introduced to the studied site where it is not naturally distributed and to replace Juniperus phoenicea(red juniper),which has low germination than the gum tree.As leguminous plants,R.monosperma and A.gummifera also provide the soil with nitrogen(Khbaya et al.1998;Vallés et al.2011),which is lacking in the sand dunes(Kachi and Hirose 1983).

The ultimate goal of this study was to investigate indigenous microbial flora that are mutualistically associated with J.phoenicea and R.monosperma,and evaluate the symbiotic associations present in the forest soil for potential exploitation in sand dune afforestation programs.Indigenous symbiotic associations with certain microflora,especially arbuscular mycorrhizal fungi,are thought to have an important effect in promoting plant development and growth.The indigenous flora have the greatest adaptability to the site conditions(Ouahmane et al.2007;Sfairi et al. 2012). Here we focused more on mycorrhizae because they represent a major rhizospheric component of most plants roots(Rillig and Mummey 2006).

Materials and methods

Study site and sampling

Field sampling was conducted in the stabilized sand dunes ecosystem of the Essaouira region in morocco,a forest of approximately 11,000 ha (CHM 2006). Several native species of some herbs and many shrubs are adapted to this habitat.The climate is hot,semi-arid sub-Mediterranean,with a mean annual precipitation of 340.6 mm and mean temperatures from 14.6 to 20°C with an average of 5 days of rain per month,and a mean daily sunshine duration of 8.2 h(HKO 2012).It is considered as a Site of Biological and Ecological Interest,code-named L25,which is a protected area by the Moroccan High Commission of Water,Forests and Fight Against Desertification(CHM 2006).

Sampling was carried out in March 2015,at the fixed dunes located in the southeast(31.2946N,9.4228W).Five repetitions of 1-3 kg of the rhizospheric soil(RS)of R.monosperma and J.phoenicea were randomly sampled between 10 and 40 cm deep.Five samples of nonrhizospheric,bare soil were also collected as a control.Samples of roots of each species were inspected for colonization,and any nodules were also collected.

Soil chemical analysis

Total organic carbon was determined using the Anne method (Pansu and Gautheyrou 2007). Total Kjeldahl nitrogen was determined with the Kjeldahl method(Egli 2008).Soil texture at different depths was determined with a Robinson pipette(Pansu and Gautheyrou 2007)and the Stokes equation. Conductivity and pH were measured using respectively a conductometer(Basic 30)and a pHmeter(Basic 20)and soil to water ratio of 1:5.Total limestone was determined with Bernard’s calcimeter,a closed setup in which excess HCl is added to soil sample to transform limestone to carbon dioxide and calculate the corresponding quantity of limestone.Available phosphorus was determined according to the Olsen(1954)using a spectrophotometer at 820 nm.

Symbiotic microbial status of Atlantic sand dunes forest

Root colonization

AM were visulaized according to the technique of Phillips and Hayman(1970).Plants roots were submerged in 10% potassium hydroxide in a water bath at 90°C for 15 min,then rinsed with dilute HCl and repeatedly with distilled water.Next,they were submerged in 5‰ trypan blue and placed in a 90°C water bath for 15 min.After repeated washing,samples were cut into 1 cm fragments,and 20 each were placed on a slide in glycerol before viewing with a light microscope.Mycorrhizal structures are stained dark blue.The following variables were selected to assess the importance of mycorrhizal colonization(Trouvelot et al.1986):

(1) Mycorrhization frequency:

where,Mfis the number of mycorrhizal fragments and Tfis the total number of fragments.

(2) Relative mycorrhization intensity M% .Root fragments are assigned a mycorrhization index from 0 to 5:0,no colonization;1,colonization of ＜1% of fragment;2,colonization of less than 10% ;3,11 to 50% ;4,51 to 90% ;5,＞91% to calculate relative intensity as:

where,n5 is the number of fragments with index 5,and so on.

(3) Absolute mycorrhization intensity

where,m is mycorrhization intensity of mycorrhized fragments.

Spore extraction

AM spores were extracted using wet sieving and decanting(Gerdemann and Nicolson 1963)and sucrose density gradient centrifugation(Walker et al.1982).Soil samples were flushed with water in a series of two sieves 800 μm and 50 μm,and the trapped fraction in each sieve was then centrifuged twice,first with water followed by vacuum filtration on Whatman filter paper,then with addition of two sucrose solutions(40% and 60% )followed by a second filtration on the same Whatman filter paper.Spores were observed and counted using a stereoscope at 40.The spores found were identified to the genus level using information provided by the websites Zor(Blaszkowski 2003)and INVAM(Morton 2017).

Mycorrhizal infectious potential(MIP)

The MIP was calculated using the most probable number(MPN) method, which was developed to estimate the density of organisms in a liquid culture.Porter(1979)was the first to use it to estimate propagules density of AM fungi in soil.It provides a relative measure of propagule density available to colonize roots.The general procedure for the MPN test is to dilute the studied soil with sterile sand for a fourfold series:1,1/4,1/16,1/64,1/256 and 1/1024.Five repetitions were made for each dilution level.Maize(Zea mays)was used as host species.Plants were grown in a greenhouse for 5 to 6 weeks to obtain a good root colonization.The plants were then harvested and the roots examined for the presence or absence of colonization using the method described above.The results are interpreted as a probability estimation of propagule number using Fisher and Yates statistical table VIII2(Fisher and Yates 1939).

Plant-growth-promoting rhizobacteria(PGPR)

PGPR nodules were collected during field sampling and at the end of the experiment,then counted and stored at 4°C for 24 h.One nodule at a time,they were surface-sterilized,rinsed and crushed in physiological water(Vincent 1970),then spread on yeast mannitol agar(YMA),the most widely used medium for rhizobia culture.Congo red was added to enhance strain differentiation(Vincent 1970).Strains appeared 3 to 5 days after incubation at 28°C.Nodules have a round,convex shape 2 to 4 mm in diameter(Somasegaran and Hoben 2012).Each colony-forming unit(CFU)suspected of being a rhizobacterium was picked and spread again on YMA for confirmation.

Soil viable microflora

Control and rhizosperic soil(RS)solutions of soils from R.monosperma and J.phoenicea were made with physiological water,the solutions were homogenized and tenfold dilutions of the solutions were spread on solid growth medium(PCA).The total number of colonies were counted after 24 h at 28°C(Foght and Aislabie 2005).Three repetitions were made for each diution.

Effects of symbiotic flora in Atlantic sand dunes forest on growth of A.gummifera and R.monosperma

The sampled RSs were mixed and divided into two;one was used in its native state,the other was sterilized at 180°C for 1 h.Seeds of R.monosperma and A.gummifera were sown in pots containing 3 kg of each type of the soils.Ten repetitions for each soil type were made to assess three growth variables:length of roots and height of shoots,number of branches,root to shoot dry mass ratio(DWr/DWs),and three nutrient variables:water,nitrogen(N)and phosphorus (P) levels. At 120 days after sowing and growth in the greenhouse,the plants were harvested and variables measured.

Statistical analysis

Results were tested according to a one-way analysis of variance(ANOVA)after validating homoscedasticity and normality using respectively Levene’s and Shapiro-Wilk tests.Tukey’s honestly significant difference(HSD)test was performed as a post hoc test.Correlations were calculated using Pearson product-moment correlation coefficient.The p-values were approximated by using the t or F distributions for the correlations.Values of p lower than 0.05 were considered significant.Statistical analyses were executed using LibreOffice Calc and R.

Results

Soil chemical analysis

The variables measured are presented in Table 1.Total organic matter in J.phoenicea RS was significantly higher than in either R.monosperma RS and the control,without any significant difference between the last two.N levels were significantly high in R.monosperma RS.There was no significant difference in N levels between J.phoenicea RS and the control.Available P,total limestone,pH,conductivity and texture,were not significantly different among the three types of soils.

Symbiotic microbial status in Atlantic sand dunes forest

Root colonization

The measured mycorrhization percentages are shown in Table 2. The mycorrhization frequency (F% ), mycorrhization intensity(M% )and mycorrhization intensity of mycorrhized fragments(m% )in J.phoenicea roots were higher than in R.monosperma roots,but the difference was not statistically significant.

Spores density and identification

Spores densities in soils are presented in Table 3.J.phoenicea RS had significantly more spores than in R.monosperma RS,and R.monosperma RS had more than in the control.Their proportions changed though.The relatively high and continuously changing portion of dark and infected spores,made it hard to keep a precise record of the proportion of each genera and to get statistically treatable information.Thus,this operation was omitted and the only global counts were retained. Ten genera among four families were found:,Acaulospora and Entrophospora(Acaulosporaceae Claroideoglomus (Claroideoglomeraceae); Gigaspora,Dentiscutata, Racocetra and Scutellospora; Gigasporaceae); Glomus, Septoglomus and Rhizophagus(Glomeraceae).

Table 2 Mycorrhization proportions for the plant species(% )

Table 3 Spore abundance in the studied soils

Table 4 Mean CFUs per gram of fresh nodule

Mycorrhizal infectious potential(MIP)

For RS of both plant species,last levels gave positive colonization.According to the MPN method and using Fisher and Yates statistical table VIII2(Fisher and Yates 1939),the number of propagules infection in both RS capable of inducing is higher than 1149×10-2propagules/g.The MPN of the control soil was 1.92×10-2propagules/g.

Plant growth promoting rhizobacteria and total viable microflora

Three nodules were collected from five RS samples from R.monosperma.After the experimental harvest,five nodules were collected from among 10 A.gummifera seedlings and two R.monosperma seedlings.The number of CFUs are given in Table 4.The number of CFUs per nodules was not significantly different between field-collected nodules and the experimental.The total CFUs for viable microflora in each soil type are presented in Table 5.CFUs were significantly higher in J.phoenicea RS than in R.monospermaRS,which was also significantly higher than in the control soil.

Table 5 Viable microorganisms counted in sampled soils

Fig.1 Lengths of shoots and roots of the two species in nonsterilized or sterilized rhizospheric soil.Different letters above the bars indicate a significant difference(ANOVA)at p ＜0.05

Symbiotic flora effects on A.gummifera and R.monosperma in Atlantic sand dunes forest

Effects on plant growth variables

The soil type did not have any significant effect on lengths of roots or shoots(Fig.1)or number of branches in either species(Fig.2).The root to shoot dry mass ratio(DWr/DW)of seedlings grown in the nonsterilized RS was significantly lower than the ratio of seedlings grown in sterilized RS(Fig.3).

Water,nitrogen and phosphorus

Fig.2 Branches number counted in the two plant species in nonsterilized or sterilized rhizospheric soil.Different letters above the bars indicate a significant difference(ANOVA)at p ＜0.05

Fig.3 Root to shoot dry mass ratio(DWr/DW)for the two plant species in nonsterilized or sterilized rhizospheric soil.Different letters above the bars indicate a significant difference(ANOVA)at p ＜0.05

Fig.4 Water content in shoots and roots of the two plant species in nonsterilized and sterilized rhizospheric soils.Different letters above the bars indicate a significant difference(ANOVA)at p ＜0.05

Water content was higher in seedlings grown in the non-RS control soil than in seedlings grown in sterilized RS.However,this difference was significant only for the R.monosperma plant roots(Fig.4).N levels were significantly different in shoots of the two species but not for R.monosperma roots.Levels were higher in the seedlings grown in the nonsterilized RS and 199% higher in R.monosperma and 700% higher in A.gummifera than in the control(Fig.5).Phosphorus levels were significantly different between the shoots and roots of both species.They were higher in seedlings grown in the nonsterilized RS(Fig.6)than in sterilized RS and control soils and higher in aerial parts than in roots in both soils,except for A.gummifera in the nonsterilized RS.Both plant species increased their P uptake up to 228% .

Fig.5 TKN in shoots and roots of the two plant species in nonsterilized and sterilized rhizospheric soils.Different letters above the bars indicate a significant difference(ANOVA)at p ＜0.05

Fig.6 Phosphorus in shoots and roots of the two plant species in nonsterilized and sterilized rhizospheric soils.Different letters above the bars indicate a significant difference(ANOVA)at p ＜0.05

Discussion

Herbaceous and woody plants are the main agents restoring soil fertility.Among the biotic factors that can promote steadier growth of plants and limit abiotic stress,arbuscular mycorrhizal symbiotic associations are the most effective for ensuring the success of revegetation strategies.

Soil chemical analysis

The most notable finding of the chemical analysis of soil was the higher percentage of organic matter in the RS of J.phoenicea.Considering the sandy texture of the soil,this level of organic matter is within the range found in most of the world’s forests.

Although the percentage organic matter in R.monosperma RS did not differ significantly from the control level,its N levels were significantly higher than in the RS of J.phoenicea and in the control.Although not significant,there was a difference in pH and total limestone between the RS for the two species compared to the control,which indicates that the plants do change the soil physicochemistry(Angers and Caron 1998).

Symbiotic microbial status of Atlantic sand dunes forest

Root colonization and spore abundance

The two studied plants have very high mycorrhization intensity and frequency.J.phoenicea has higher mycorrhization percentage compared to R.monosperma,although the roots of the latter are less sclerified and easier to penetrate than those of J.phoenicea.It is matter of affinity and preference between mycorrhizal fungi and different host plants(Su et al.2011).R.monosperma can have endophytic fungi(Bacon and Hinton 1996;Sieber and Grünig 2006)in its roots and aerial parts(Mack and Rudgers 2008),which may compete with mycorrhizal fungi for carbon and potentially limit but not prevent mycorrhization.The plant still needs other benefits such as water supply and protection that endophytic fungi cannot provide.

The number of spores in rhizospheric soils of the two plants is very important.This abundance can be explained by the season when sampling was performed.In fact,Smith(1980)showed that the spore density is highest in the spring,and others have also shown seasonal variation in the spore abundance(Phillips and Hayman 1970;Sieber and Grünig 2006).The difference between the numbers of spores enumerated can be explained by differences in the root systems;J.phoenicia has an extremely ramified root system,thus providing a greater sporulation area compared to R.monosperma root system,which is pivoted and less branched.Lower AM spores numbers have been noted by several authors for dunes in Italy;Florida,United States(Sylvia and Will 1988); Lake Huron, North America(Koske et al.1975);Rhode Island,United States(Koske and Halvorson,1981)and New South Wales,Australia(Koske,1975,as cited by B?aszkowski and Czerniawska 2011).Large numbers can also be explained by the method of counting. In this study, all spores were counted,including the dead and infected ones.The spore number presented by the control soil can be explained by the continuous mobility that marks the dune geosystem.The degree of dune stability also influences the mycorrhizal population(Giovannetti and Nicolson 1983);spores initially produced on the root surface and around the roots are moved in several ways to bare soil,and after death and/or deterioration of the root system that produces these spores,the soil becomes easy to mobilize by wind or runoff during rainy periods.

AM and soil contents of N and P

The AM are known for low tolerance to high concentrations of phosphorus in the soil.When the level of available phosphorus exceeds 140 ppm,the percentage of infection decrease(Amijee et al.1989;Koide and Li 1990).Mycorrhizal development is optimal when phosphorus levels in the soil are around 50 ppm.In the studied dunes,phosphorus levels were very low,as in dunes generally(Koske and Poison 2011),and ranged between 8 and 14 ppm.In such cases, mycorrhizae play the role of phosphorus prospector in the soil,concentrating P at the hyphosphere and mycorrhizosphere(Rambelli 1973)and storing it in the mycelium and storage organs,then distributing it to the plant(Schachtman et al.1998).In general,native AM increase the levels of N and P for the seedlings(Ouahmane et al.2012).Soil N level for the two plants species differed significantly;the RS of R.monosperma had higher levels than the RS of J.phoenicea and was negatively correlated with mycorrhization percentage for R. monosperma(r=-0.76,p ＜0.05)but weak and nonsignificant for J.phoenicea(r=-0.55,p=0.12),which can be explained by the presence of rhizobacteria,a limiting factor through their competition with mycorrhizal fungi.C/N ratios indicate that R. monosperma supplies soils with nitrogen,whereas the C/N ratio for J.phoenicea is much higher,which indicates consumption of N.However,there was no apparent correlation between the C/N ratio and degree of mycorrhizal infection (R. monosperma: r=0.0005,p ＜0.01;J.phoenicea:r=0.01 p ＜0.05),which goes along with the trophic independence of mycorrhizal fungi toward carbon other than that provided by the plant.

Mycorrhizal infectious potential

The MIP presented by the rhizospheric soils of both species compared with that for the control is very important.Propagules can be spores,vesicles,free totipotent mycelium or mycelium attached to the colonized roots,which have the ability to germinate and infect.In dunes,mycorrhizae are known to have a very important infectious potential.

Inoculation test

Among plant variables,only root and shoot lengths of R.monosperma differed significantly from those in nonsterilized RS. For both plant species,the DWr/DW ratio showed an important and significant difference between the sterilized RS and nonsterilized RS. This ratio is an indicator of the resource availability and stress state for the plants.Roots are directly influenced by low soil availability of either water or nutrients,which leads to greater DWr/DW ratios.In this experiment,seedlings that were grown in the sterilized RS had a higher ratio than those in the nonsterilized RS.Roots of the latter series did not have to expand their root system,and thus biomass,as much as those in the sterilized RS.

Water levels differed significantly only at the root level.Phosphorus levels showed an important significant difference between the two series,between both species and between the roots and shoots of the same species of the same series,except between roots and shoots of A.gummifera in the nonsterilized RS.Total Kjeldahl nitrogen was also significantly different between the two series except for R.monosperma roots,but the reason is not clear.Both species belong to the Fabaceae family,meaning they both are likely to have nodules on their roots,which was confirmed with direct in situ observation. Yet, after the experiment,five nodules were collected from A.gummifera series(50% incidence)and only two from the R.monosperma series(20% incidence),while three nodules were field-collected from R. monosperma (60% incidence).Without any statistical analyses of incidence,we can only speculate that R. monosperma has a latent period for nodulation.

Conclusion

This study showed that soil fertility was enhanced by the considered plants species. J. phoenicea enhanced soil organic matter,while R.monosperma and A.gummifera enhanced soil nitrogen levels.Also,a diverse and abundant mycorrhizae flora was found,and its infectious potential is worth considering in agricultural and silvicultural practices. The significant amelioration of the growth and nutrition variables in experimental conditions prove the important role of mycorrhizal fungi for the plants.Mycorrhizae not only help plants stabilize the sands,but they also help transform the soil into a live and livable habitat.