Accepted Manuscript

Title: Versatile use of Azospirillum brasilense strains tagged

with egfp and mCherry genes for the visualization of biofilms
associated with wheat roots

Authors: Alberto Ramirez-Mata, Miguel Ramales Pacheco,

Saul Jijon Moreno, Maria Luisa Xiqui-Vazquez, Beatriz E.
Baca

PII: S0944-5013(18)30327-6
DOI: https://doi.org/10.1016/j.micres.2018.07.007
Reference: MICRES 26192

To appear in:

Received date: 20-3-2018

Revised date: 10-7-2018
Accepted date: 18-7-2018

Please cite this article as: Ramirez-Mata A, Pacheco MR, Moreno SJ, Xiqui-Vazquez
ML, Baca BE, Versatile use of Azospirillum brasilense strains tagged with egfp
and mCherry genes for the visualization of biofilms associated with wheat roots,
Microbiological Research (2018), https://doi.org/10.1016/j.micres.2018.07.007

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Versatile use of Azospirillum brasilense strains tagged with egfp and

mCherry genes for the visualization of biofilms associated with wheat

roots

PT
Alberto Ramirez-Mata, Miguel Ramales Pacheco, Saul Jijon Moreno, Maria Luisa Xiqui-

Vazquez, and Beatriz E. Baca*.

RI
Centro de Investigaciones en Ciencias Microbiologicas, Benemerita Universidad Autonoma

SC
de Puebla. Edif. IC11, Ciudad Universitaria, Puebla, Puebla 72570, Mexico

U
* Corresponding author: Beatriz Eugenia Baca
N
E-mail beatriz.baca@correo.buap.mx, beatrizebaca@gmail.com
A
M
D
TE

ABSTRACT

This study reports the introduction of egfp or mCherry markers to the Sp245, Sp7, and M40
EP

wild-type strains of Azospirillum brasilense and the hhkB (encoding for a putative hybrid

histidine kinase) minus mutant an isogenic strain of A. brasilense Sp245 to monitor

CC

colonization of wheat (Triticum aestivum). Two plasmids were constructed: (1) the pJMS-2
A

suicide plasmid derived from pSUP202 and harboring the mCherry gene expressed under

the constitutive kanamycin resistance promoter to create a cis tag and (2) the broad-range

plasmid pMP2449-5 that carries the mCherry gene under the lac promoter, which is derived

from the plasmid pMP2444; to create the in trans tag. The stability of the plasmids

1
encoding egfp and mCherry were confirmed in vitro for seven days of bacterial growth, and

then, the A. brasilense strains harboring the plasmids were studied under nonselective

conditions for adherence to seeds and, at seven or 14 days post-inoculation, for wheat root

colonization. The utility of the labeled strains was proven by observation, using

PT
fluorescence microscopy and confocal laser scanning microscopy (CLSM) in wheat plants

inoculated with the labeled strains and compared with the CFU g–1 for seed and wheat root.

RI
The method was suitable for observation of the in situ formation of mini-colonies, enabled

SC
visualization of bacterial colonization sites on large root fragments, and showed adherence

to germinated seeds and root colonization of all strains by cell counts and direct

U
microscopic examination. Thus, we are able to quantify the structures of the biofilms

formed by each strain. N

A
M

Key words: Azospirillum, biofilm formation, mCherry fluorescent protein, enhanced-green

fluorescent protein
D
TE

1. Introduction
EP

Among plant-beneficial microorganisms, plant growth-promoting rhizobacteria (PGPR)

colonize plant root systems and enhance plant growth and nutrition through a variety of
CC

mechanisms, including direct effects on nutrient uptake (e.g., N2-fixation, P-mobilization,

and iron chelation), and enhance root growth through the production of phytohormones and
A

nitric oxide production (Bashan and de Bashan, 2010). Azospirillum brasilense is one of

many PGPRs and is known to have a broad host plant range (Casanovas et al., 2015). It has

been isolated from the rhizospheres of wheat, other crops, and several other agronomic

2
plants (Malinich and Bauer, 2018; Vacheron et al., 2013). Some strains exhibit additional

traits that can directly or indirectly benefit the health and yield of plants. Direct growth

stimulation can be mediated by facilitated acquisition of nutrients or by modulation of

phytohormone equilibrium. The indirect effect stems from increased plant resistance to

PT
diseases (Tortora et al., 2011; Spaepen et al., 2014). This resistance can result from the

induction of the defense system of the plant or from antagonism shown by A. brasilense

RI
toward plant pathogens. The effective colonization of plant roots by A. brasilense is a

SC
prerequisite of this important role in growth promotion, irrespective of the mechanism of

action (Bashan and de Bashan, 2010). A chemotaxis mechanism toward organic acids,

U
sugars, some amino acids, and root exudates facilitate Azospirillum colonization of the

N
roots of plants, which depends on motility by the polar flagellum (Alexandre et al., 2000).
A
The initial attachment of Azospirillum to roots is also dependent on the use of the polar
M

flagellum, which allows the bacteria to adsorb to the roots (Croes et al., 1993; Malinich and

Bauer, 2018). Azospirillum has also been observed to transition into nonmotile, cyst-like
D

cells in aerated cultures, inducing cell-to-cell clumping and flocculation, and promoting an
TE

ordered set of changes in cell envelope properties and metabolic activities (Sadasivan and
EP

Neyra, 1985; Bible et al., 2015). If stress conditions persist, the bacterium implements a

variety of additional strategies, such as biofilm formation. It is now common knowledge

CC

that bacteria in natural environments survive by forming biofilms (Römling et al., 2013).

Biofilms are highly structured, surface-attached communities of cells encased in a self-

A

produced extracellular matrix, which is relevant to the ability of cells to colonize and

establish within the rhizospheres of host plants. Microcolonies and microaggregates of A.

brasilense cells form during the colonization of plant root surfaces, and strains unable to

undergo flocculation are deficient in colonization (Pereg-Gerk et al., 1998). Thus, it is

3
essential to develop methodological strategies to analyze in situ colonization without

affecting the structures formed by bacteria and root tissues. Studies of Azospirillum

colonization and localization on inoculated root surfaces were performed with A. brasilense

Sp7 and A. brasilense Sp245 strains using strain-specific oligonucleotide probes to localize

PT
different A. brasilense strains via fluorescence in situ hybridization (FISH) in combination

with CLSM (Assmus et al., 1995). Detection of reporter gene expression by Azospirillum

RI
on plant roots also allows researchers to estimate the number of bacteria on roots inoculated

SC
by detecting the enzymatic activity of beta-galactosidase or beta-glucuronidase produced by

strains carrying lacZ or gusA genes (Arsène et al., 1994; Vande Broek et al., 1993;

U
Schenberger-Santos et al., 2017).

N
Other researchers have used genes encoding autofluorescent proteins, such as egfp and
A
mCherry, to monitor colonization of Pseudomonas fluorescens, Rhizobium leguminosarum
M

bv. viciae RBL5560, or Mesorhizobium loti R7A, to plants (Stuurman et. al., 2000;

Bloemberg et al., 2000; Lagendijk et al., 2010). Their major advantage for the use of GFP
D

and mCherry proteins are their independence of substrate or energy reserves for detection
TE

of fluorescence (Lagendijk et al., 2010). However, the main drawback of eGFP is that it
EP

seems to be very stable, which in turn renders the protein less valuable for studies of

transient (real-time) gene expression (Tombolini et al., 1997).

CC

Our objective was to develop a set of genetic tools for tagging A. brasilense strains

with egfp and mCherry genes expressed under the lac promoter and either harbored in high
A

copy number of plasmids or constitutively expressed under the constitutive kanamycin

promoter from a chromosomal integration site; each of them could be maintained in the cell

without adding antibiotics and were expressed at a level that allows visualization. In

4
addition, we showed that this method is not deleterious for A. brasilense growth and our

constructs can be used just as successfully in different isolates of A. brasilense strains.

2. Material and methods

PT
2.1. Bacterial strains, plasmids, and culturing conditions

RI
The bacterial strains and plasmids used in this study are listed in Table 1. All

SC
cloning and transferring tests were conducted with Escherichia coli DH5α and E. coli S17.1

strains. The E. coli strains were routinely grown at 37 °C in Luria–Bertani broth (LB, Life

U
Technologies, Grand Island, NY). When necessary, the following antibiotics were added

N
(final concentration): ampicillin (Ap; 50 µg/mL); kanamycin (Km; 25 µg/mL); and
A
tetracycline (Tc; 10 µg/mL). The A. brasilense strains were grown at 30 °C in K-malate
M

minimal medium, LB* (LB modified as follows: 10 mM NaCl; 2.5 mM MgSO4, and 2.5

mM CaCl2) or Congo Red medium (CR) (Jijón-Moreno et al., 2015), and antibiotics were
D

added at the following concentrations when required: Km, 50 µg/mL; Tc, 15 µg/mL; or
TE

gentamycin, (Gm) 15 µg/mL.

EP

2.2. DNA manipulation

CC

The chromosomal DNA of A. brasilense was isolated, as described previously

(Jijón-Moreno et al., 2015). Standard techniques were used for enzymatic restriction of
A

DNA, agarose gel electrophoresis, isolation of DNA fragments from low-melting-point

agarose gels, and ligation (Sambrook and Russell, 2001). Restriction fragments were

purified from agarose gels using the ZymocleanTM ZR-96 Gel DNA Recovery Kits (Zymo

Research, Tustin, California USA), depending on the size of the fragment. Transformation
5
of bacterial cells with plasmid DNA was performed, as described previously (Carreño-

López et al., 2009). Polymerase chain reactions (PCRs) were performed in a Bio-Rad

Laboratories iCycler thermal cycler, using High Fidelity Platinum Taq DNA Polymerase

(Invitrogen) for gene cloning, and Taq DNA recombinant polymerase (Invitrogen) for

PT
diagnostic reactions, following the amplification protocols detailed previously (Jijón-

Moreno et al., 2015; Carreño-López et al., 2009). All constructs involving PCR techniques,

RI
mutants obtained, and transferred cells were verified by sequence analysis. Nucleotide

SC
sequences were determined using universal and custom oligonucleotide primers from the

Sequencing Unit of the National Autonomous University of Mexico (UNAM).

U
N
2.3. Construction of egfp- and mCherry-based reporter plasmids and tagged Azospirillum
A
strains
M

The pJMS-2 (ΔhisC1::pkmCherry-aph-hisC1) plasmid and the plasmid control

ΔhisC1::aph named pJMS-1 are suicide plasmids derived from pSUP202 (Simon et al.,
D

1983) that contain the mCherry gene downstream of the promoter (pkm) of the aph gene,
TE

encoding KmR followed by the kanamycin resistance gene integrated into the chromosomal
EP

ΔhisC1 gene of the A. brasilense Sp7, Sp245, and M40 strains. All construction steps are

detailed in the Supplementary Material (Fig. 1S-ab). The plasmids pJMS-2 and pJMS-1
CC

allow the insertion of mCherry as a single copy into the chromosomes of A. brasilense

strains through double homologous recombination into the hisC1 gene encoding the
A

enzyme histidinol phosphate aminotransferase, mutagenized with a deletion (ΔhisC1),

(Castro-Guerrero et al., 2012). The partial deletion of hisC1 gene in tagged strains was

confirmed by PCR and amplicon sequencing. The pJMS-2 and pJMS-1 plasmids were

transferred to all A. brasilense strains, using E. coli S17.1 (pJMS-2, ΔhisC1pkmCherryKmR

6
) or E. coli S17.1 (pJMS-1, ΔhisC1KmR) as donors by conjugation to A. brasilense strains,

as previously described (Carreño-López et al., 2009).

The plasmids pMP2449-5 and pMP2449-3 were constructed by substituting the egfp

gene with the mCherry gene; both are plasmids derived from the broad host range

PT
pMP2444 plasmid (Bloemberg et al., 2000). The constructions of both plasmids are shown

in the Supplementary Material (Fig. 2S-a). The plasmid pMP2444 and its derivatives

RI
constructed in this study were transferred to each of the A. brasilense strains using E. coli

SC
S17.1 (pMP2449-5), E. coli S17.1 (pMP2449-3), and E. coli S17.1 (pMP2444) as donors

by conjugation, as previously described (Carreño-López et al., 2009).

U
2.4. In vitro plasmid stability N
A
An analysis of plasmid stability was performed in A. brasilense Sp7, Sp245, and
M

M40 strains and the A. brasilense ΔhhkB isogenic mutant (A. brasilense C556) from Sp245

that harbor the plasmids pMP2444, pMP2449-5, and pMP2449-3. The strains were grown
D

to stationary phase in K-malate medium without adding antibiotics, and 10 μL of each

TE

strain culture was inoculated into new tubes every 24 h, diluted 1000-fold in the same
EP

medium, and then grown exponentially. All stability assays were initiated by the dilution of

cells to 105 colony-forming units (CFU/mL) in nonselective medium followed by growth

CC

overnight. Cultures were again diluted and grown overnight in nonselective medium (as

described above), and this procedure was repeated for seven days. After each dilution,
A

aliquots were plated on CR agar medium without antibiotics. The colonies were inoculated

on K-malate medium supplemented with Gm at the concentration indicated in the culture

conditions section to determine whether the plasmid was missing. All experiments were

repeated twice. In addition, the corresponding plasmids were isolated, as described in

7
Sambrook and Russell (2001). The fluorescence analysis from the tagged strains (Table 1)

with eGFP or mCherry proteins was performed in K-malate medium after growing for 24,

48 or 72 h. All strains were visualized by fluorescence microscopy in an inverted (TE 2000-

U Eclipse) Nikon microscope, using an oil-objective 100 × 1.25. The images were captured

PT
with a Nikon (DS-QilMc) camera and edited using the NIS elements in Nikon software.

RI
Table 1. Bacterial strains and plasmids.

SC
2.5. Adherence of A. brasilense strains to seeds and plant growth conditions during

U
inoculation

N
To test adherence to seeds and perform plant inoculations, Azospirillum strains were
A
grown in LB* without antibiotics to the early stationary growth phase (i.e., to an optical
M

density at 600 nm [OD600] of 1.2). Cultures were diluted 1:1000 in K-malate medium

grown at 30 °C for 18 h with agitation at 150 rpm; then the cells were diluted to OD600 of
D

0.125 in 10 µL, which corresponds to approximately 107 CFU/mL. Each inoculum was
TE

confirmed by the determination of CFU/mL on solid CR medium, supplemented with the

EP

corresponding antibiotic.

Seeds of the wheat (Triticum aestivum) variety ‘Nana’ were obtained from the
CC

Campo Agrícola Experimental del Valle de México (INIFAP). To eliminate contaminating

microorganisms, approximately 0.05 g of seeds (200 seeds per 100 mL) was treated with
A

1% (v/v) Tween 20 (Sigma Aldrich) in sterile deionized water for 2 min and washed three

times with sterile deionized water. Subsequently, the seeds were incubated in sodium

hypochlorite 1% (v/v) with agitation (60 rpm) for 30 min. Then, seeds were rinsed eight

times with demineralized sterile water (5 min per rinse). The seeds were immersed in 40
8
mL of a solution containing cycloheximide 150 µg/mL, streptomycin 250 µg/mL,

tetracycline 20 µg/mL, and fluconazole 150 µg/mL, with agitation at 60 rpm for 10 min. At

the end of the disinfection process, the seeds were rinsed five times with demineralized

sterile water. Then, 15 seeds were inoculated in a 125 mL flask with 10 mL of K-malate

PT
medium containing approximately 107 CFU/mL of each strain and maintained at room

temperature with shaking (60 rpm) for 30 min. Afterward, the inoculated seeds were

RI
carefully washed twice with a phosphate-buffered saline (PBS, pH 6.8) solution. One

SC
milliter of PBS (pH 6.8) was added and the bacteria adhered to the seeds were sonicated for

10 min (Bransonic Ultrasonics 1200; 50/60 Hz, 117 volts, 0.7 amps, Danbury, CT).

U
Adhered cells were determined by growth in a plate containing CR medium supplemented

with the corresponding antibiotic. N

A
To determine the colonization of wheat by A. brasilense Sp7-02, A. brasilense
M

Sp245 (pMP2444), and A. brasilense M40-02, the seeds were treated as above and then

placed on Petri plates filled with agar 0.6% (w/v) in darkness for two days. Germinated
D

seeds were transferred into sterile glass tubes (20 × 150 mm) containing 15 mL of
TE

Hoagland hydroponic solution supplemented with 25 mM KNO3, as described by Arsène et

EP

al. (1994). The tubes contained a cut sterile Eppendorf tube to support the seedlings and

were covered by additional test tubes that were larger in diameter (30 × 200 mm). For
CC

seven days, they were maintained in a growth chamber at 28 °C under a daylight period of

12 h with 70% relative humidity. The plants (nine plants in each of three independent
A

experiments) were inoculated with approximately 107 CFU/mL of the A. brasilense Sp7-02,

M40-02, or Sp245 (pMP2444) strains and then were maintained under the same conditions

in a growth chamber for seven days for Sp7-02 and Sp245 (pMP2444) or 14 days for M40-

02. After seven or 14 days post-inoculation (dpi), the plants were removed from the system
9
and carefully washed with a PBS (pH 6.8) solution. Colonizing bacteria were determined

by colony counting, as described above. The roots were subsequently analyzed for the

presence of bacteria and biofilms, using confocal laser scanning microscopy (CLSM), and

the images were edited using the NIS elements in Nikon software.

PT
2.6. Microscopic visualization of bacteria in the free-living state by fluorescence and in

RI
plant roots using CLSM

SC
To visualize single egfp-tagged or mCherry-tagged bacterial cells, overnight

cultures were centrifuged, resuspended in PBS (pH 6.8), spread on slides, air dried, and

U
subsequently mounted under coverslips in PBS. Fluorescence microscopy used a Nikon

N
imaging fluorescence microscope equipped with appropriate filter sets: eGFP excitation
A
was set at 488 nm and mCherry excitation was set at 587 nm. For the visualization of egfp
M

or mCherry, tagged A. brasilense Sp7, Sp245, M40, C556 (ΔhhkB) mutant, and the

respective negative controls of each strain were examined for fluorescent cells; photos were
D

taken and analyzed with software. For visualization after colonization of wheat roots and
TE

the formation of biofilms by the Sp245, M40, M40-01, and M40-02 strains, freshly
EP

harvested whole roots or hand-cut root slices were placed on a glass slide. To prepare the

longitudinal slices, the roots were placed on a glass slide and cut into halves with a razor
CC

blade. Whole roots or slices were embedded in PBS, and eGFP-fluorescence was visualized

in the cells. Both fluorophores were detected using an inverted CLSM, equipped with four
A

lasers: 405 nm, 488 nm, 561 nm, and 640 nm fitted with a 10X (Numerical aperture NA

0.45), 20X (NA 0.75), or 60X (oil TIRF NA 1.49) objective. eGFP fluorescent cells were

visualized using an excitation wavelength of 488 nm with fluorescence emission captured

between 505 nm and 530 nm and mCherry fluorescent cells were visualized using an
10
excitation wavelength of 561 nm with fluorescence emission captured between 585 nm and

615 nm. Image analysis was carried out using the standard software supplied with the

microscope or the software ImageJ.

PT
2.7. Statistical analysis

A two-tailed Student’s t-test was used to find significance, with significance set at

RI
0.001 to assess the variability of the experimental data from nine inoculated seeds or plants.

SC
Fig. 1.

U
Fig. 2. N
A
M

3. Results

3.1. Visualization of A. brasilense Sp7, Sp245, and M40 strains and the C556 ΔhhkB
D

mutant tagged with eGFP and mCherry fluorescent proteins

TE

All A. brasilense strains under study were successfully transferred with the different
EP

plasmids harboring egfp or mCherry genes, resulting in stable expression of eGFP and

mCherry proteins in the free-living state (Fig. 1). The stability of each plasmid was
CC

confirmed in vitro for seven days of bacterial growth and after 30 min of cell contact with

seeds (Figs. 1 and 2). In fact, A. brasilense tagged strains were shown to actively adhere to
A

wheat seeds, forming a significant surface population after 30 min of inoculation, ranging

from 104 to 106 CFU/mL (Fig. 2). In addition, eGFP and mCherry-expressing A. brasilense

strains had colonized plant root at seven or 14 days after roots inoculation under

nonselective conditions (Figs. 3, 4, and 5). In addition, a plasmid was used to label cells in
11
a chromosomal locus, the hisC1 gene, to assess the expression of fluorescence from

mCherry proteins, under the constitutive kanamycin-resistant promoter. As a general

observation, A. brasilense strains containing plasmids harboring the gene encoding the

mCherry protein (Figs. 1 and 3) emitted a similar fluorescence to that of cells labeled with a

PT
chromosomal insertion (Figs. 1, 3, and 5). However, the latter construction required 72 h to

exhibit mCherry expression. Although the plasmid vector pBBR1 is a multi-copy plasmid

RI
(Fang and Helinski, 1991), the A. brasilense strains harboring the plasmid did not show

SC
reduced growth, compared to that of the corresponding wild-types (Supplementary

Material, Fig. 2S-b). The method of tagging A. brasilense strains in this study was shown to

U
be useful and very stable under nonselective conditions.

N
It was also demonstrated that eGFP carried in pMP2444 is an excellent reporter for
A
A. brasilense Sp245 visualizing bacteria (Figs. 1 and 3) after 24 and 72 h of growth without
M

antibiotics in the stationary phase; the result in the free-living state is shown in Fig. 1.

Interestingly, after seven days under hydroponic growth conditions, and no selective
D

pressure, plants inoculated with A. brasilense Sp245 (pMP2444) or A. brasilense Sp7-

TE

(pMP2444)-derived strains labeled with eGFP showed stable, detectable, and strong
EP

fluorescence (Fig. 4 and video in the Supplementary Material). Comparable colonization of

surface plant cells by all strains was observed, and the bacteria formed cellular aggregates,
CC

microcolonies, and biofilms (Figs. 3 and 4). As described by Karatan and Watnick (2009),

biofilms exist under different structures, such as the monolayer where bacteria are
A

distributed on the surface as a single layer or as microcolonies (Lebeaux and Ghigo, 2012).

Multilayer biofilm, multiple layers of bacteria encased in an extracellular matrix, and

aggregates, which could be formed by a specific signal (Bible et al., 2015). An

amplification of the image of roots showed that the bacteria formed aggregates on root hairs
12
and formed microcolonies (Fig. 4). Some motile bacteria were also observed (Video in the

Supplementary Material), which suggests that the cells were actively growing and

synthesizing stable eGFP. In addition, biofilm was formed, mostly present on the upper part

of the root; the A. brasilense Sp245 strain preferred colonizing this root section rather than

PT
the root hairs (Fig. 4 and video in the Supplementary Material). This result was confirmed

by cell counting, which showed 7.3 × 1011 CFU g–1 of plant root fresh weight. A similar

RI
result was obtained with Sp7 (pMP2444), as shown in Supplementary Material (video),

SC
indicating extensive wheat root colonization.

U
Fig. 3.

N
A
Because it is often desirable to detect two different fluorescent proteins (FPs)
M

simultaneously, the solution was to express the mCherry protein under the KmR constitutive

promoter, which was cloned into the chromosome by placing it at a native gene location
D

(hisC1 gene), using homologous recombination. These labeled A. brasilense strains

TE

constitutively expressed the mCherry protein from a stable chromosomal locality (Figs. 1,
EP

3, and 5). Derived strains with tags in the chromosomal locus hisC1 were previously found

not to decrease cell growth, as shown in the Supplementary Material (Fig. 2S-b), likely
CC

because A. brasilense contains at least two loci coding for aromatic amino acyl transferases,

as demonstrated by Castro-Guerrero et al. (2012) and Jijón-Moreno et al. (2015). The

A

generation of mCherry-tagged A. brasilense strains M40-02, Sp7-02, and Sp245-02, and a

mutant in a putative hybrid histidine kinase HhkB isogenic to the Sp245 strain (the

construction of which is described in detail elsewhere) was a very suitable tool to visualize

the colonization of wheat inoculated with this fluorescent protein, especially when plant
13
autofluorescence is strong (Figs. 3 and 5). Furthermore, colonization at seven days post-

inoculation was confirmed by cell counting of A. brasilense Sp7-02, obtaining 6.4 × 109

CFU g–1 of plant root fresh weight, under hydroponic conditions without selection pressure,

with reduced cell counts compared to those obtained with A. brasilense Sp245 (pMP2444).

PT
Fig. 4.

RI
SC
A. brasilense M40-02 was found to be a stable construct suitable for the analysis of

colonization at 14 days post-inoculation. Although A. brasilense M40-02, which is derived

U
from the M40 strain isolated from maize in Argentina (Garcia de Salamone et al., 1996),

N
was able to actively colonize wheat roots, as demonstrated by re-isolation experiments; the
A
cells were found to be firmly attached, with 1.5 × 109 CFU g–1 of wheat root fresh weight.
M

The A. brasilense M40-02 strain was able to actively and highly colonize wheat roots under

hydroponic conditions without antibiotic selective pressure (Fig. 5), the where M40-02
D

strain develops microcolonies and biofilms all along the rhizoplane from the middle to the
TE

root cap of wheat (Fig. 5, Panels A–D). Biofilm formation was measured, showing a width
EP

of 213.83 μm, a height of 213.84 μm, and a depth of 65.17 μm (Fig. 5, Panel D).

Fig. 5.
CC

4. Discussion
A

Our aim was to implement new reporter assays to visualize eGFP or mCherry

labeled bacteria with CLSM as an effective, fast, and noninvasive tool that allows analysis

of bacteria-plant interactions, while preserving the integrity of the organisms under study

because specific staining procedures are not required for their application. We developed
14
microscopic observations by CLSM as an experimental approach that allows monitoring of

colonization levels. The Sp7 and Sp245 strains were found to efficiently and thoroughly

adhere to seeds or colonize root surfaces as previously demonstrated (Assmus et al., 1995),

based on specific tagged oligonucleotides. In addition, the fluorescence pattern on plants of

PT
strains tagged with eGFP corresponds to the sites of lateral root emergence and root tips

using histological staining (Arsène et al., 1994; Vande Broek et al., 1993; Santos et al.,

RI
2017). However, the use of the genes encoding FPs, such as egfp or mCherry, as reporters

SC
was useful because their fluorescence does not require any cofactors for expression,

enabling detection via nondestructive sampling (Lagendijk et al., 2010). The advantage of

U
plasmid encoding FPs is that they are present in multiple copies; this might, under certain

N
conditions, improve the production of FPs and does not disrupt host genes by chromosomal
A
integration. In addition, the plasmids used were very stably maintained in all A. brasilense
M

strains and the derivative mutant; cells growth was not reduced as antibiotics were not

added to the hydroponic solution. After re-isolation from seeds or roots, we found the
D

original strains at a high level, as shown by cell counts; these levels were comparable to
TE

those described in previous studies of A. brasilense Sp7 or M40 strains (Garcia de

EP

Salamone et al.,1996; Greer-Phillips et al., 2004; Compant et al., 2010). However, in the

case of adherence, all tagged strains showed less CFU/mL, likely from a metabolic burden
CC

to the cell, which resulted in a growth delay over a short time (30 min).

The use of chromosomal insertion of mCherry in the hisC gene was also useful, and
A

no change in growth was obtained compared to the respective wild-type strains. In addition,

this insertion allowed an improved evaluation of fluorescence by avoiding conditions in

which the autofluorescence of root plants is very strong (Fig. 5). The plasmid harboring the

mCherry gene was stable independent of strain used. This method provides a system that is
15
genetically stable during free-living, seed adherence, and colonizing wheat roots (Figs. 1, 3,

4, and 5). Under hydroponic conditions, biofilm formation was determined, allowing an

easy to use tool for further detection of genes likely involved in biofilm formation in

Azospirillum-plant interactions. Biofilms might provide an ecological advantage for

PT
surviving environmental perturbations because they create a protective niche against other

microorganisms found in soil.

RI
We compared the fluorescence of strains tagged in cis in a chromosomal locus and

SC
in plasmids encoding both FPs expressed under constitutive promoters and found that the

fluorescence emitted in inoculated plants was similar among the tested strains.

U
In summary, the tagged strains proved stable for at least 120 generations in the free-

N
living state, as well as after the colonization of wheat roots under nonselective conditions at
A
7 or 14 dpi. Moreover, no damage was observed in the inoculated plants, making these
M

egfp-tagged or mCherry-tagged strains useful for the study of the molecular mechanisms

involved in plant attachment and colonization using mutant strains. Our results indicate that
D

these tools may be useful for monitoring colonization by different A. brasilense wild-type
TE

strains and an isogenic mutant of Sp245 strain isolated from different plants.
EP

5. Conclusions
CC

The method developed is suitable for in situ analysis of biofilm production and

phenotypes of mutants with altered colonization traits for a comprehensive understanding

A

of the features in Azospirillum that are involved in plant colonization. The method is an

elegant approach to follow the behavior and activity of bacteria on plant root surfaces

before extensive use in field applications as a biofertilizer or for bioremediation.

16
Acknowledgments

The authors thank Rosa R. Mouriño Pérez at Centro de Investigación Científica y de

Educación Superior de Ensenada (CICESE) for providing the pJV15.2 plasmid; Inés García

de Salamone at the Facultad de Agronomía, Universidad de Buenos Aires for providing the

PT
A. brasilense M40 strain; and René Hortelano Santa Rosa at INIFAP for providing the

wheat seeds. This study was funded by the Consejo Nacional de Ciencia y Tecnología

RI
(CONACYT grant INFR-2014-01-225923). The vice-rector of investigations and the vice-

SC
rector of graduate studies (VIEP) supported the initial experiments. We thank the reviewers

for their valuable comments.

U
N
A
Conflict of interest
M

All authors declare no conflict of interest

D
TE

References
EP

Alexandre, G., Greer, S.E., Zhulin, I.B., 2000. Energy taxis is the dominant behavior in
CC

Azospirillum brasilense. J. Bacteriol.182, 6042–6048.

A

Arsène, F., Katupitiya, S., Kennedy, I.R., Elmerich, C., 1994. Use of lac Z fusions to

study the expression of nif genes of Azospirillum brasilense in association with

plants. Mol. Plant-Microbe Interact.7, 748–757.

17
 Assmus, B., Hutzler, P., Kirchhof, G., et al., 1995, In situ localization of Azospirillum

brasilense in the rhizosphere of wheat with fluorescently labeled, rRNA-

targeted oligonucleotide probes and scanning confocal laser microscopy. Appl.

Environ. Microbiol. 61, 1013–1019.

PT
Baldani, V.L.D., Alvarez, M.A., Baldani, J.I., Döbereiner, J., 1986. Establishment of

inoculated Azospirillum spp. in the rhizosphere and in roots of field grown wheat

RI
and sorghum. Plant Soil 90, 35–46.

SC
Bashan, Y., de-Bashan, L.E., 2010. How the plant growth-promoting bacterium

Azospirillum promotes plant growth – a critical assessment. Advance

U
Agronomy. 108, pp 77–136.
N
A
Bible, A.N., Khalsa-Moyers, G.K., Mukherjee, T. et al., 2015. Metabolic adaptations of
M

Azospirillum brasilense to oxygen stress by cell-to-cell clumping and

flocculation. Appl. Environ. Microbiol. 81, 8346-8357.

D

Bloemberg, G.V., Wijfles, A.H.M., Lamers, G.E.M., et al., 2000. Simultaneous imaging
TE

of Pseudomonas fluorescens WCS365 populations expressing three different

EP

autofluorescent proteins in the rhizosphere: New perspectives for studying

microbial communities. Mol. Plant-Microbe Interact. 13, 1170–1176.

CC

Carreño-López, R., Sánchez, A., Camargo, N., et. al., 2009. Characterization of chsA, a
A

new gene controlling the chemotactic response, in Azospirillum brasilense Sp7.

Arch. Microbiol. 191, 501-507.

18
 Casanovas, E.M., Fasciglione, G., Barassi, C.A., 2015. Azospirillum spp. and related

PGPRs inocula use in intensive agriculture. Handbook for Azospirillum. DOI

10.1007/978-3-319-06542-7-25.

Castro-Guerrero, J., Romero, A., Aguilar, J.J., Xiqui, M.L. et al., 2012. The hisC1 gene,

PT
encoding aromatic amino acid aminotransferase-1 in Azospirillum brasilense

Sp7, expressed in wheat. Plant Soil. 356, 139–150.

RI
Compant, S., Clément, C., Sessitsch A., 2010. Plant growth-promoting bacteria in

SC
rhizo- and endosphere of plants: Their role, colonization, mechanisms involved

U
and prospects for utilization. Soil Biol. Biochem. 42, 669-678.

N
Croes, C., Moens, S., van Bastelaere, E., et. al., 1993. The polar flagellum mediates
A
Azospirillum brasilense adsorption to wheat roots. J. Gen. Microbiol. 139,
M

2261–2269.

Fang, F.C., Helinski, D.R., 1991. Broad-host-range properties of plasmid-RK2 –

D
TE

importance of overlapping genes encoding the plasmid replication initiation

protein TrfA. J. Bacteriol. 173, 5861–5868.

EP

Garcia de Salamone, I.E., Döbereiner, J., et. al., 1996. Biological nitrogen fixation in

Azospirillum strain-maize genotype associations as evaluated by the a 15N

CC

isotope dilution technique. Biology Fertile Soils. 23, 249-256.

A

Greer-Phillips, S.E., Stephens, B.B., Alexandre, G., 2004. An energy taxis transducer

promotes root colonization by Azospirillum brasilense. J. Bacteriol. 186, 6595-

6004.

19
 Jijón-Moreno, S., Marcos-Jiménez, C., Pedraza, R.O. et al., 2015. The ipdC, hisC1 and

hisC2 genes involved in indole-3-acetic acid production used as alternative

phylogenetic markers in Azospirillum brasilense. Antonie van Leeuwenhoek J.

107, 1501-1517. DOI 10.1007/s10482-015-0444-0.

PT
Karatan, E., Watnick, P. 2009. Signals, regulatory networks, and material that built and

break bacterial biofilm. Microbiol. Mol Biol. Rev. 2, 310-347.

RI
Lagendijk, E.L., Validov, S., Lamers, G.E.M., et al., 2010. Genetic tools for tagging

SC
Gram-negative bacteria with mCherry for visualization in vitro and in natural

U
habitats, biofilm and pathogenicity studies. FEMS Microbiol. Let. 305, 81–90.

N
Lebeaux, D., Ghigo JM. 2012. Infections associées aux biofilms. Quelles perspectives
A
thérapeutiques issues de la recherche fondamentale ?Medicine/Sciences. 28;
M

727-739.
D

Malinich, E.A., Bauer, C.E., 2018. The plant growth promoting bacterium Azospirillum
TE

brasilense is vertically transmitted in Phaseolus vulgaris (common bean).

Symbiosis https://DOI.org/10.1007/s13199-018-0539-2
EP

Pereg-Gerk, L., Paquelin, A., Gounon, P., et al., 1998. A transcriptional regulator of the

LuxR-UhpA family, FlcA, controls flocculation and wheat root surface

CC

colonization by Azospirillum brasilense Sp7. Mol. Plant-Microbe Interact. 11,

A

177–187.

Römling, U., Galperin, M.Y., Gomelsky, M. 2013. Cyclic di-GMP: the first 25 years of

a universal bacterial second messenger. Microbiol Mol Biol Rev. 77, 1–52.

20
 Sadasivan L., Neyra C.A. 1985. Flocculation in Azospirillum brasilense and

Azospirillum lipoferum: Exopolysaccharides and cyst formation. J. Bacteriol.

163, 716-723.

Sambrook, J., Russell, D.W., 2001. Molecular cloning: a laboratory manual. 3th edn.

PT
CSHL Press, New York, USA.

RI
Schenberger-Santos, A.R.S., Etto, R.M., Furmam, R.W., et al., 2017. Labeled

Azospirillum brasilense wild type and excretion-ammonium strains in

SC
association with barley roots. Plant Physiol. Biochem. 118, 422-426. DOI:

U
10.1016/j.plaphy.2017.07.003.

N
Simon, R., Priefer, U., Pühler, A., 1983. A broad host range mobilization system for in
A
vivo genetic engineering: transposon mutagenesis in gram negative bacteria.
M

Biotechnology. 1, 784–790.
D

Spaepen, S., Bossuyt, S., Engelen, K., et al., 2014. Phenotypical and molecular
TE

responses of Arabidopsis thaliana roots as a result of inoculation with the auxin-

producing bacterium Azospirillum brasilense. New Phytologist J. 201, 850–886.

EP

Stuurman N., Pacios Bras, C., Schlaman, H.R.M. et. al., 2000. Use of green fluorescent

protein color variants expressed on stable broad-host-range vectors to visualize

CC

rhizobia interacting with plants. Mol. Plant-Microbe Interact. 13, 1163-1169.

A

Tarrand, J.J., Krieg, N.R. Döbereiner, J., 1978 A taxonomic study of the Spirillum

lipoferum group, with the description of a new genus, Azospirillum gen. nov.

and two species, Azospirillum lipoferum (Beijerinck) comb. nov. and

Azospirillum brasilense sp. nov. Can. J. Microbiol. 24, 967–980.

21
 Tombolini, R., Unge, A., Davey, M.E. et al., 1997. Flow cytometric and microscopic

analysis of GFP-tagged Pseudomonas fluorescens bacteria. FEMS Microbiol.

Ecol. 22, 17–28.

Tortora, M.L., Díaz-Ricci, J.C., Pedraza, R.O., 2011. Protection of strawberry plants

PT
(Fragaria ananassa Duch.) against anthracnose disease induced by Azospirillum

brasilense. Plant Soil. DOI 10.1007/s11104-011-0916-6.

RI
Vacheron, J., Desbrosses, G., Bouffaud, M.L., et al., 2013. Plant growth-promoting

SC
rhizobacteria and root system functioning. Frontiers in Plant Sci. 4, 356.

U
Vande-Broek, A., Michiels, J., Van Gool, A., Vanderleyden, J., 1993. Spatial-temporal

N
colonization patterns of Azospirillum brasilense on the wheat root surface and
A
expression of the bacterial nif H gene during association. Mol. Plant-Microbe
M

Interact. 6, 592–600.
D

Verdín J., Bartnicki-Garcia S., Riquelme M., 2009. Functional stratification of the
TE

Spitzenkörper of Neurospora crassa. Mol. Microbiol. 74, 1044-1053.

EP

Figure legends

Fig. 1. Microscopic visualization of tagged bacteria in the free-living state. All strains were
CC

harvested at 48 h of growth, as described in Materials and Methods section. Fluorescence

A

images were taken with an exposure time of 1.2 sec for all strains to compare the

differences in brightness between the strains, with excitation at 587 nm and emission at 610

nm for mCherry protein and excitation at 488 nm and emission at 509 nm for eGFP

22
protein. The images were made with fluorescence microscopy and differential interference

contrast (DIC) images (same experiment) by light microscopy. Scale bar = 50 μm.

Fig. 2. Adherence determination of wheat seeds inoculated with Azospirillum brasilense

PT
tagged strains. Wheat seeds (Triticum aestivum var. Nana) were disinfected, as described in

Materials and Methods section. Each of the nine seeds was inoculated for 30 min. with

RI
approximately 107 CFU/mL of each A. brasilense tagged strain grown in K-malate medium

SC
for 18 h at 30 °C with shaking at 60 rpm. Lanes: 1. Sp245, 2. C556, 3. Sp245 (pMP2449-5),

4. C556 (pMP2449-5), 5. Sp245 (pMP2449-3), 6. CC6 (pMP2449-3), 7. M40, 8. M40-01,

U
9. M40-02. Data are the result of three independent experiments with nine seeds each

N
(0.05/g sec weight). *represents a p-value of <0.001.
A
M

Fig. 3. Visualization of the root of Triticum aestivum var. Nana inoculated with

Azospirillum brasilense derivative strains tagged with mCherry fluorescent protein.

D

Visualization of surface colonization of wheat roots with each strain at 14 dpi. Seedlings
TE

were inoculated with A. brasilense M40-02, Sp7-02, and Sp245-02 (approximately 107
EP

CFU/mL), in which the mCherry gene was chromosomally inserted in the hisC1 gene, as

described in the Materials and Methods section. The Sp7 strain was used as a negative
CC

control. White arrows indicate motile cells or biofilm formation. Scale bar = 50 μm.
A

Fig. 4. Visualization of roots of Triticum aestivum var. Nana inoculated with Azospirillum

brasilense Sp245 (pMP2444). The root was observed at seven days after inoculation using

an inverted CLSM. Root colonization by the A. brasilense Sp245 strain was observed in the

root elongation zone in the Nana genotype. Panel A. Image obtained with a Plan Apo
23
lambda 10× objective. Panel B. Image obtained with a Plan Apo lambda 20× objective.

Panel C. Image obtained with a Plan Apo lambda 60× objective. All images were edited

with NIS confocal and ImageJ software. White arrows indicate that bacteria are attached to

the root cell surface, forming microcolonies and biofilm; free cells are visible near the

PT
upper root hairs. Magnification of the region indicated by the arrow shows high

colonization of the root hair spaces. Images were obtained from a Nikon CLSM C2+. Scale

RI
bar = 50 µm.

SC
Fig. 5. Visualization of roots of Triticum aestivum var. Nana inoculated with Azospirillum

U
brasilense M40-02 derivative strain tagged with mCherry fluorescent protein. Visualization

N
of surface colonization of wheat roots with corresponding strains is shown. Seedlings were
A
inoculated with A. brasilense M40-02 (approximately 107 CFU/mL), in which the mCherry
M

gene was chromosomally inserted in the hisC1 gene, as described in the Materials and

Methods section. Sections were visualized 14 days after inoculation and microscopically
D

viewed for mCherry fluorescence. mCherry-expressing cells growing on roots are shown by
TE

arrows. High surface root colonization, microcolonies, the formation of biofilm, and free
EP

cells are also shown. Panel A. Composite image of large regions of wheat roots inoculated

with A. brasilense M40-02 for analysis. Images were collected by confocal microscopy
CC

with a Plan Apo lambda 20× objective and edited with NIS elements and ImageJ software.

Panel B. Merged image section collected by confocal microscopy with an Apo TIRF 60×
A

oil objective. Bar = 50 µm. Panel C. Merged image section collected by confocal

microscopy with a Plan Apo lambda 20× objective. Panel D. Three-dimensional

reconstruction image to evaluate biofilm formation on the wheat root surface. The images

were collected by confocal microscopy with a Plan Apo lambda 20× objective. Images
24
were obtained from a Nikon CLSM C2+ objective Plan Apo TIRF 60× oil and edited with

software NIS confocal elements. Scale bar = 50 µm.

PT
RI
SC
U
N
A
M
D
TE
EP
CC
A

25
 A
CC
EP
TE
D
M
A
N
U
SC
RI

26
PT
 A
CC
EP
TE
D
M
A
N
U
SC
RI

27
PT
 A
CC
EP
TE
D
M
A
N
U
SC
RI

28
PT
 A
CC
EP
TE
D
M
A
N
U
SC
RI

29
PT
 PT
RI
SC
Table 1. Bacterial strains and plasmids.

U
Strains and Relevant characteristics References/source

N
Plasmids

A
Strains M
Escherichia coli
E. coli DH5α endA1 hsdR17 supE44 thi-1 λ- recA1 gyrA96 relA1 ΔlacU169 Ø80 (ΔlacZΔM15) Bio-Rad Laboratories
E. coli S17.1 pro, thi, hsd:: RP4-2-Tc:: Mu-Km Tn7, SmR SpR ATCC® 47055TM Simon et al. 1983
D

Azospirillum brasilense
Sp7 Wild-type Tarrand et al. 1978
TE

Sp245 Wild-type Baldani et al. 1986

M40 Wild-type Garcia de Salamone et al. 1996
Sp7 (pMP2444) Sp7 harboring pMP2444, GmR This study
Sp245(pMP2444) Sp245 harboring pMP2444, GmR This study
EP

M40 (pMP2444) M40 harboring pMP2444, GmR This study

Sp7 (pMP2449-5) Sp7 harboring pMP2449-5, GmR This study
Sp245(pMP2449-5) Sp245 harboring pMP2449-5, GmR This study
CC

M40 (pMP2449-5) M40 harboring pMP2440-5, GmR This study

C556(pMP2449-5) Isogenic from Sp245, which was mutated in hhkB gene harboring pMP2449-5, GmR This study
Sp7 (pMP2449-3) Sp7 harboring pMP2449-3, GmR This study
Sp245(pMP2449-3) Sp245 harboring pMP2449-3, GmR This study
A

M40 (pMP2449-3) M40 harboring pMP2440-3, GmR This study

C556 (pMP2449-3) Isogenic from Sp245, which was mutated in hhkB gene harboring pMP2449-3, GmR This study
Sp7-01 Sp7 aph gene inserted in hisC1 gene, KmR This study
Sp245-01 Sp245 aph gene inserted in hisC1 gene, KmR This study
M40-01 M40 aph gene inserted in hisC1 gene, KmR This study
Sp7-02 Sp7 tagged strain with the mCherry gene under pkm expression inserted in hisC1 gene This study
Sp245-02 Sp245 tagged strain with the mCherry gene under pkm expression and inserted in hisC1 This study

30
 PT
RI
SC
gene This study
M40-02 M40 tagged strain with the mCherry gene under pkm expression and inserted in hisC1
gene
Plasmids Simon et al. 1983

U
pSUP202 Cloning vector ApR, CmR, TcR
pKD4 Plasmid red recombinase, ApR, KmR Datsenko and Wanner, 2000

N
pBSK+ Cloning vector ApR Stratagene

A
pBSK-mCherry Cloning vector, containing mCherry gene, ApR This study
pGEM-Teasy Cloning vector ApR Promega
pMP2444
M
Plasmid broad host range GmR, containing egfp expressed under control of lac Bloemberg et al. 2000
promoter, cloned in pBBR-1
pJMS-2 Suicide plasmid derivative of pSUP202, harboring a fragment of 5.29 kb containing This study
hisC1, pkmR, mCherry, aph, hisC1 with ApR, KmR, CmR
D

pJMS-1 Suicide plasmid derivative of pSUP202, harboring a fragment of 4.46 kb containing This study
hisC1, kmR, hisC1 with ApR, KmR, CmR
TE

pMP2449-5 Plasmid harboring a fragment of 724 bp from mCherry gene expressed under control of
lac promoter, cloned in pBBR-1 This study
pMP2449-3 Plasmid harboring a fragment of 724 bp from mCherry gene cloned in 3′ direction of lac This study
promoter, cloned in pBBR-1
EP

Ap = ampicillin; Gm = gentamicin; Tc = tetracycline, Km = kanamycin.

CC
A

31