A new medium for the detection of fluorescent pigment production by pseudomonads

Authors


E-mail: lamichhane@unitus.it

Abstract

Two media (King’s B [KB] and CSGA) commonly used for the detection of fluorescent pigment by Pseudomonas spp. were compared to a new medium proposed in this study, PGS agar. Thirty-nine strains of 10 different species of Pseudomonas from several geographic regions were screened. The efficacies of these media were examined under several conditions, including the addition of iron-binding substances or supplementation with extra iron. The medium developed, which included an iron-binding agent, was the most permissive for production of fluorescent pigments when compared to KB and CSGA. Thirty-seven of the 39 pseudomonad strains screened were highly fluorescent on this new medium compared to 15 and 16 strains, respectively, on KB or CSGA. The optimal composition of the medium per litre was Bacto peptone 10 g, gelatin 20 g, sucrose 20 g, agar 15 g, dipotassium hydrogen phosphate 1 g, magnesium sulphate heptahydrate 1 g and conalbumin 2 g. Protocol validation tests performed through an intra-laboratory study in comparison to KB demonstrated the effectiveness of the new PGS medium.

Introduction

Most of the bacterial species belonging to the genus Pseudomonas are well known for their capacity to produce siderophores that fluoresce under UV light at 365 nm. Recently, there has been an increased interest in the Pseudomonas spp. because of their ability to cause diseases on a large number of hosts and because of the possible use of siderophores as biopesticides (Wilson, 1997). The siderophores are chelating agents with specific natural low molecular weights that transport iron (III) to the cell surface in the form of a complex. Iron (III) is released from this complex during its reductive assimilation (Kamnev et al., 2000). In fluorescent pseudomonads, pyoverdines, a group of structurally related fluorescent siderophores, represent the primary iron uptake system. In addition, many species can also synthesize other siderophores, such as pyochelin and quinolobactin, or can acquire iron bound to a variety of exogenous chelators, including many heterologous siderophores (Cornelis & Matthijs, 2002; Poole & McKay, 2003; Schalk, 2008). However, the ability of these bacteria to excrete the fluorescent pigments when grown on synthetic media differs and often depends on the composition of the growth medium, as described by many authors (Burton et al., 1947; King et al., 1948; Palumbo, 1973; Jones et al., 1986).

The production of fluorescent pigments (FP) is affected by many compounds and increases in the presence of some molecules such as sodium, potassium and magnesium, phosphate and sulphates (Georgia & Poe, 1931; Bhattacharya, 2010). In addition, the quantities of iron present in a given medium significantly influence the FP excretion. In fact, the restricted availability of iron, one of the most essential microelements for living cells, has been reported as the main nutritional condition controlling the biosynthesis and excretion of fluorescent pigments (Totter & Moseley, 1953; Jones et al., 1986).

Generally, King’s medium B (KB) described by King et al. (1954) is widely used to detect the fluorescent pigments produced by Pseudomonas spp. However, many strains fail to produce or enhance the pigment excretion when grown on this medium and many authors have attempted to optimize conditions for Pseudomonas spp. pigment synthesis (Gouda & Greppin, 1965; Luisetti et al., 1972; Reyes et al., 1981). Furthermore, numerous attempts have been made to enhance FP production using different iron-binding substances (IBS), such as the egg white protein conalbumin (synonym ovo-transferrin), as chelating substances to render the media iron-free by sequestering metallic contaminants (Warner & Weber, 1953; Garibaldi, 1960, 1967; Dulla et al., 2010).

This paper describes the preparation of a new diagnostic medium, based on peptone, sucrose and salts, demonstrating how the addition of IBS to a certain medium makes it effective for the detection of FP production. The aim of this work was to develop an alternative medium to KB in which almost all the pseudomonads, and in particular Pseudomonas syringae pv. actinidiae, can produce FP.

Materials and methods

Bacterial strains

Six fluorescent and four non-fluorescent phytopathogenic pseudomonads grown on KB medium were used in this study. The fluorescent species used for the investigation were Pseudomonas syringae pv. syringae (Pss), P. syringae pv. lachrymans (Psl), P. syringae pv. aptata (Psap), P. savastanoi pv. savastanoi (Psav), P. viridiflava (Pv) and P. avellanae (Pa) and the non-fluorescent strains were P. syringae pv. theae (Pst), P. syringae pv. mendocina (Psm), P. corrugata (Pc) and P. syringae pv. actinidiae (Psa). One strain for each species was used except in the cases of P. avellanae and P. syringae pv. actinidiae where four and 27 strains, respectively, of different origins were used. Most of the reference strains used in this study were from the National Collection of Plant Pathogenic Bacteria (NCPPB, UK), the International Collection of Micro-organisms from Plants (ICMP, New Zealand), the French Collection of Plant Pathogenic Bacteria (CFBP, France), and the Fruit Tree Research Center (CRA-FRU, Italy). The remaining strains were provided directly by the researchers (Table 1).

Table 1. Reference strains of phytopathogenic Pseudomonas spp. used in this study
BacteriaaStrainFluorescence on King’s BHostCountrySource
  1. aPseudomonas syringae pv. syringae (Pss), P. syringae pv. theae (Pst), P. syringae pv. lachrymans (Psl), P. syringae pv. aptata (Psap), P. syringae pv. mendocina (Psm), P. savastanoi pv. savastanoi (Psav), P. viridiflava (Pv), P. corrugata (Pc), P. avellanae (Pa) and P. syringae pv. actinidiae (Psa).

Pss3909 B+ Actinidia chinensis ItalyA. Calzolari
Pst2598 Camellia sinensis JapanNCPPB
Psl6458+ Cucumis melo USACFBP
PsapDPP36+ Cucumis melo ItalyL. Varvaro
PsmDPP26 Populus nigra ItalyL. Varvaro
PsavPseNE 107+ Olea europaea NepalG. M. Balestra
Pv4254a+ Actinidia chinensis ItalyG. M. Balestra
Pc2445 Solanum lycopersicum UKNCPPB
Pa3489+ Corylus avellana GreeceNCPPB
3491+ Corylus avellana GreeceNCPPB
4223+ Corylus avellana ItalyNCPPB
4226+ Corylus avellana ItalyNCPPB
PsaPsa V Actinidia sp.ChinaL. Huang
Psa VI Actinidia sp.ChinaL. Huang
Psa VII Actinidia sp.ChinaL. Huang
36.43+ Actinidia sp.FranceF. Poliakoff
36.44 Actinidia sp.FranceF. Poliakoff
36.45 Actinidia sp.FranceF. Poliakoff
36.46 Actinidia sp.FranceF. Poliakoff
18839 Actinidia sp.New ZealandICMP
18875 Actinidia sp.New ZealandICMP
18804+ Actinidia sp.New ZealandICMP
18882+ Actinidia sp.New ZealandICMP
Psa KN2 Actinidia sp.South KoreaY. J. Koh
CJW7+ Actinidia sp.South KoreaY. J. Koh
WGE 12+ Actinidia sp.South KoreaY. J. Koh
YCJ2-1+ Actinidia sp.South KoreaY. J. Koh
Psa Por 352 Actinidia sp.PortugalG. M. Balestra
Psa Por 354 Actinidia sp.PortugalG. M. Balestra
Psa Por 356 Actinidia sp.PortugalG. M. Balestra
38.17 Actinidia sp.SwitzerlandF. Poliakoff
Psa 827 Actinidia sp.SpainG. M. Balestra
Psa 829 Actinidia sp.SpainG. M. Balestra
CRA-Fru 12.50 Actinidia sp.ItalyCRA-Fru
CRA-Fru 12.51 Actinidia sp.ItalyCRA-Fru
CRA-Fru 13.02 Actinidia sp.ItalyCRA-Fru
Psa 782 Actinidia sp.ItalyCFBP
Psa 7285 Actinidia sp.ItalyCFBP
Psa 829 Actinidia sp.ItalyCFBP

Choice of ingredients

The lack of FP production on KB by some pseudomonads was associated with the presence of iron in the medium. For this reason a medium was prepared using different carbon and nitrogen sources and salts that had positive effects on bacterial growth and the consequent FP production (Georgia & Poe, 1931, 1932). The basic ingredients used to prepare the medium were Bacto peptone, gelatin, sucrose, dipotassium hydrogen phosphate, magnesium sulphate and agar (PGS).

Iron-binding substances

The following chelating agents were used in order to prepare the iron-free PGS agar medium.

Egg white

Egg whites from fresh eggs were separated in sterile conditions by slightly modifying the method described by Garibaldi (1960). Eggs were carefully washed under flooding tap water, dried and immersed in 70% ethanol for 5 min. The ethanol was discarded and its residue was removed by flaming the shell very carefully. The egg shell was cracked and the egg white was separated in a sterile container, vortexed and stored at −20°C until used. Before use, the egg white was thawed at room temperature and then placed in a 45°C waterbath, vortexed and added to the medium.

Conalbumin

Conalbumin (Sigma-Aldrich) is one of the major types of albumin found in chicken egg whites that is commercially available in pure crystalline form. This protein has different metal combining properties, including iron (Warner & Weber, 1953). Different concentrations of conalbumin solutions were added to the medium by microfiltration. The protein was not subjected to excessive heating, as it is heat sensitive.

Tannic acid

Besides egg white and conalbumin, which are known for their iron-binding capacity (Garibaldi, 1967), tannic acid was also used, which binds ferric ions by influencing quorum sensing in Pseudomonas spp. (Dulla et al., 2010). Although the iron-binding properties of tannic acid were known, no previous study exists on the use of this substance for the enhancement of fluorescence.

Ferric iron

One of the most important elements that influences fluorescence production is iron (Totter & Moseley, 1953). The influence of iron availability on fluorescent pigment excretion was evaluated by adding different concentrations of ferric iron to the media.

Preliminary tests to determine the concentration of ingredients

The type and concentration of carbon sources present in a given medium are very important for the excretion of FP by a given Pseudomonas sp. Jones et al. (1986) found that the ability of some phytopathogenic pseudomonads to fluoresce on iron-deficient media depended on the type and concentration of carbon sources in the media. In addition, the authors confirmed that the production of FP occurs at a high rate on media containing sucrose. In preliminary studies here, media were prepared with and without peptone, gelatin and sucrose, one at a time, to ascertain if the missing ingredient affected FP production. The concentrations were also varied (0, 1, 2, 3, 4 and 5% w/v) during optimization. The same procedure was adopted for the IBS concentrations, and only the minimum and maximum concentrations that gave significant differences were selected for the final test. These were: egg white, 10 and 20%; conalbumin, 0·002 and 0·004%; FeCl3, 5 and 50 μm; and tannic acid, 5 and 50 μm. The performances of KB and CSGA media were compared with and without an iron supplement in order to evaluate its influence. Finally, the same quantity of ferric iron and IBS, except the egg white (ratio 1:1), were used in the media to verify whether the supplemented iron can be saturated by the addition of the same quantity of tannic acid.

Media composition and preparation of experimental plates

The compositions (per litre) of the media used in this study were: (i) KB: proteose peptone (Difco) 20 g, dipotassium hydrogen phosphate 1·5 g, magnesium sulphate heptahydrate 1·5 g, glycerol 15 mL, agar 15 g; (ii) CSGA: casaminic acid (vitamin free, Difco) 10 g, sucrose 10 g, dipotassium hydrogen phosphate 1 g, magnesium sulphate heptahydrate 1 g, gelatin 30 g, agar 20 g; and (iii) PGS agar: Bacto peptone (Difco) 10 g, gelatin 20 g, sucrose 20 g, dipotassium hydrogen phosphate 1 g, magnesium sulphate heptahydrate 1 g and agar 15 g. All the ingredients were dissolved by heating until boiling and autoclaved at 121°C for 15 min, after which the substrate was left to cool to 55°C. The incorporation of egg white was made by simple mixing; however, conalbumin and tannic acid were added by microfiltration (0·22 μm) after dissolving them in water. Finally, the media were poured into Petri dishes under sterile conditions. Media without chelating agents were used as controls in the experiments (Table 2).

Table 2. Fluorescent pigment excretion by phytopathogenic Pseudomonas spp. strains on different media supplemented with different substances
MediumaPssPstPslPsapPsmPsavPvPcPaPsa
  1. a(I) 10% egg white; (II) 20% egg white; (III) 0·002% conalbumin; (IV) 0·004% conalbumin; (V) 5 μm ferric chloride; (VI) 50 μm ferric chloride; (VII) 5 μm tannic acid; (VIII) 50 μm tannic acid; and (IX) 50 + 50 μm ferric chloride and iron-binding substances (1:1 ratio).

  2. b++++ very rich fluorescence; +++ rich fluorescence; ++ good fluorescence; + weak fluorescence; − no fluorescence.

  3. cReduction of fluorescent pigment production due to poor growth.

  4. dThe pigment colour changes from blue to green.

  5. eGreek strains appear fluorescent only after 72 h.

  6. fThe production of yellow pigment is highly enhanced.

  7. gSome strains from France, Korea and New Zealand are positive.

KB++b++++++++++eg
KB+ I++++++++++++++++++++++
KB+ II+c,d+c,d+c,d+c,d+c,d+c,d+c,d+c
KB+ III++++++++++++++++++++++
KB+ IV++c++c++c++c++c++c++c+c
KB+ V
KB+ VI
KB+ VII+++++++++++++++++++++
KB+ VIII+c+c+c+c+c+c+c+c
KB+ IX+++++++++++++++
CSGA++++++++g
CSGA+ I+++++++++++++++++++++++
CSGA+ II++++++++++++++++
CSGA+ III+++++++++++++++++++++++
CSGA+ IV+++++++++++++++
CSGA+ V
CSGA+ VI
CSGA+ VII++++++++++++++++++++++
CSGA+ VIII++++++++
CSGA+ IX+++++++++++++++
PGSf
PGS+ I++++++++++++++++++++++++++++++++
PGS+ II++++++++++++++++++++++
PGS+ III++++++++++++++++++++++++++++++++
PGS+ IV++++++++++++++++++++++
PGS+ V
PGS+ VI
PGS+ VII+++++++++++++++++++++
PGS+ VIII+++++++++++++++
PGS+ IX+++++++++++++

Both growth and FP production were evaluated on these media with and without an IBS supplement. The performance of the PGS agar medium was compared with KB and CSGA, media commonly used for the detection of FP (King et al., 1954; Luisetti et al., 1972).

Inoculation of experimental plates

All the phytopathogenic bacterial strains were first grown on M9 minimal medium (Schaad et al., 2001), and plates containing KB, CSGA and PGS agar with and without addition of the IBS were inoculated with a fresh 24 h bacterial culture. Five replicates for each bacterial strain for each medium were made. Inoculated plates were incubated at 26 ± 1°C.

Detection of fluorescence

Plates were examined at 24, 48 and 72 h after incubation under UV light at 360 nm (Sylvania Blacklite Blue Tubes F 15T8-BLB). KB and CSGA, two media especially designed for the detection of fluorescence, were used as controls. The FP production on these agar media were compared with the production on these media supplemented with different ingredients at different percentages (Table 2). In addition, the two media were compared with the PGS agar medium supplemented with the same ingredients in the same proportions.

Efficacy of experimental plates supplemented with egg white

To verify whether the medium supplemented with egg white maintained its efficacy over time, plates were stored both at room temperature (approximately 20°C) and at 4°C. Inoculations on these plates were made on days 5, 10, 15, 20 and 30 after storage to detect the excretion of FP. KB was used as a control.

Protocol validation study

Three important parameters, relative accuracy (AC), relative sensitivity (SE) and relative specificity (SP) were calculated for the validation of the medium proposed in this study (Olivier et al., 2010).

These parameters were evaluated by using the following formulae: (i) relative sensitivity (SE) = 100PA/(ND + PA), where PA and ND stand for positive agreement and negative deviation, respectively; (ii) relative specificity (SP) = 100NA/(NA + PD), where NA and PD stand for negative agreement and positive deviation, respectively; and (iii) relative accuracy (AC) = 100(PA + NA)/(NA + PA + PD + ND).

The investigation was carried out by using an intra-laboratory study processing the same samples and comparing the results obtained by different lab technicians. Three lab technicians took part in the intra-laboratory comparison of this test procedure and each of them received 20 samples to analyse. The samples included 20 bacterial strains belonging to different families, 10 fluorescent and 10 non-fluorescent, on KB medium (Table 1). Eighteen of these strains, half of them fluorescent and half non-fluorescent, on KB medium, were numbered randomly and used anonymously. The name of the species was provided for the positive and negative controls. The bacteria were streaked on KB and PGS agar media and incubated at 26 ± 1°C. The FP production was inspected under UV light on days 1, 2 and 3 on both media, and the results were compared (Table 4).

Results

The optimal composition (per litre) of PGS agar medium supplemented with IBS, on which all the tested phytopathogenic Pseudomonas spp. produced an excellent fluorescent pigment, was the following: Bacto peptone 10 g, gelatin 20 g, sucrose 20 g, agar 15 g, dipotassium hydrogen phosphate 1 g, magnesium sulphate heptahydrate 1 g supplemented with 10% egg white, 0·002% conalbumin or 5 μm tannic acid.

The FP production on KB, CSGA and PGS agar with or without addition of IBS differed significantly. The pseudomonad strains positive on KB were also positive on CSGA medium. Pseudomonas syringae pv. theae, which did not fluoresce on KB medium, did excrete the FP on CSGA. In contrast, none of the tested bacteria produced FP on PGS agar without IBS supplements (Table 2). The concentration of peptone, gelatin and sucrose greatly influenced the effectiveness of the medium for the detection of FP. Almost all the bacteria produced very poor FP when one of these ingredients was omitted. However, FP production increased proportionately with increasing concentrations of peptone, gelatin and sucrose up to 0·01, 0·02 and 0·02% (w/v), respectively. At higher concentrations the result remained unchanged (data not shown).

The FP produced by Pseudomonas spp. increased on KB and CSGA media supplemented with IBS compared to the media with no supplements. The fluorescence did not occur on PGS agar without the addition of IBS, but significant increases were observed once IBS were incorporated. The optimal supplemental quantities for maximum FP excretion were 10% egg white (w/v), 0·002% conalbumin (w/v) or 5 μm tannic acid (Table 2). The diagnostic media lost their capacity to excrete FP once conalbumin was completely saturated with iron. FP were not produced when the concentration of ferric chloride was ≥5 μm, but were recovered on media containing iron after an additional IBS supplement of the same amount (Table 2). The results showed that the conalbumin, both from fresh eggs or in purified commercial form, with its well-established property to chelate and restrict the availability of iron, allowed enhancement of FP production on media.

The IBS supplementation significantly affected the growth of the bacteria (Fig. 1). There was a reduction in growth as the amount of these substances increased in the media. In fact, the lower fluorescence observed with higher IBS supplement quantities were due to the decreased growth. Media supplemented with egg whites at percentages greater than 15% significantly reduced the bacterial growth and consequently the excretion of FP (Table 2). The three media used in this study, when supplemented with the same quantity of IBS, showed different levels of FP production. The addition of an optimal amount of these substances greatly stimulated the excretion of FP without any interference to growth. In terms of enhancement in fluorescence, PGS agar gave the best response, followed by CSGA and KB (Fig. 2; Table 2).

Figure 1.

 Growth of pseudomonad species on media supplemented with different iron-binding substances after 36 h of incubation at 26°C. (a) King’s medium B (KB); (b) KB supplemented with 10% egg white, (c) 5 μm ferric chloride, or (d) 50 μm ferric chloride; (e) PGS agar without supplementation; (f) KB supplemented with 5 μm tannic acid; (g) 50 μm tannic acid or (h) 50 μm ferric chloride and tannic acid. Clockwise from upper left in each plate: Pseudomonas syringae pv. syringae; Pseudomonas syringae pv. theae; Pseudomonas syringae pv. mendocina and Pseudomonas corrugata.

Figure 2.

 Examples of pigment excretion by Pseudomonas syringae pv. actinidiae on different media under ultraviolet light (365 nm) after 36 h of incubation at 26°C. (a,b) King’s medium B (KB) with and without a 0·002% conalbumin supplement; (c,d) CSGA medium with and without a 0·002% conalbumin supplement; and (e,f) PGS agar medium with and without a 0·002% conalbumin supplement. Clockwise from the upper left in each plate: strain Psa V (China); strain 36.43 (France); strain 18839 (New Zealand); and strain Psa 827 (Spain).

The FP excreted on the media with restricted iron availability was significantly greater than the media where no iron-binding agents were supplemented (Fig. 2). The levels of FP detected from 39 strains belonging to different Pseudomonas spp. showed that the new PGS agar proposed in this study when supplemented with 10% egg white or 0·002% conalbumin is the best diagnostic medium. Among the fluorescent strains used, the pigment excretion on PGS agar supplemented with conalbumin was superior to FP observed on KB and CSGA. Of 27 strains of Pseudomonas syringae pv. actinidiae, only six produced FP on KB and CSGA, whereas they all showed a positive response on PGS agar supplemented with IBS. The non-fluorescent strains of P. syringae pv. mendocina and P. corrugata did not excrete FP on any of these media (Table 3). Some differences in fluorescence were observed at 24, 48 and 72 h among the media. The level of FP production was highest within 24 h on PGS agar medium for most of the species. On KB and CSGA media the optimal level was not reached until closer to 48 h. Surprisingly, Greek isolates of P. avellanae fluoresced only after 72 h of incubation on KB.

Table 3. Fluorescent pigment production by Pseudomonas spp. on three different media
BacteriaaNo. strains testedFluorescence on
KBCSGAPGS agar
  1. aPseudomonas syringae pv. syringae (Pss), P. syringae pv. theae (Pst), P. syringae pv. lachrymans (Psl), P. syringae pv. aptata (Psap), P. syringae pv. mendocina (Psm), P. savastanoi pv. savastanoi (Psav), P. viridiflava (Pv), P. corrugata (Pc), P. avellanae (Pa) and P. syringae pv. actinidiae (Psa).

  2. bStrains 36.43 from France; 18804 and 18882 from New Zealand and CJW7, WGE 12 and YCJ2-1 from Korea are positive.

Pss1111
Pst1011
Psl1111
Psap1111
Psm1000
Psav1111
Pv1111
Pc1000
Pa4444
Psa276b6b27

The validation study showed very clear differences in the capacity of KB and PGS agar media to support FP production and confirmed that the latter is far more effective. The number of fluorescent strains observed on the media and their respective values of relative sensitivity, relative specificity and relative accuracy differed significantly over time. On KB medium, the best results were obtained after 2 days (Table 4). After this time point they started to decrease. The results were even more interesting on PGS agar, where the optimal values were observed within 1 day and decreased afterwards (Table 4). Finally, no difference was observed among the plates stored at 4°C and those stored at room temperature, as the production of FP was observed consistently on both for 1 month after preparation (data not shown).

Table 4. Fluorescence test procedure: comparison between expected results and positive, negative and doubtful obtained results per sample (in number) on two different media; determination of performance criteria
No. of positive results
MediumFluorescence after daysExpectedObtainedPositive agreement (PA)Negative deviation (ND)Relative sensitivity (SE)%
  1. aThe doubtful results were considered as negative results.

KB13014141646·67  
23016161453·33  
33010102033·33  
PGS agar1514242982·35  
25140401178·43  
35139391276·47  
No. of negative results
MediumFluorescence after daysExpectedObtainedObtained doubtfula resultsNegative agreement (NA)Positive deviation (PD)Relative specifity (SP)%Relative accuracy (AC)%
KB13046130010073·33
23044330010076·67
33044630010066·67
PGS agar191809010085·00
292039010081·67
392159010080·00

Discussion

The concentration of iron in the growth medium is the key factor that controls the excretion of FP by Pseudomonas spp. (Totter & Moseley, 1953). Because iron is an essential element for the synthesis of the cytochrome system enzymes, it cannot be removed completely from the media. The use of IBS on the one hand enables the production of FP and on the other makes iron available to bacteria. However, the quantity of a particular IBS, such as conalbumin, to be used should carefully consider the iron content of the medium (1 mL of egg white is equivalent to 1·7 mg of conalbumin and binds approximately 21 μg of Fe). Conalbumin in stoichiometric excess of iron inhibits the growth of Gram-positive microorganisms and may affect Gram-negative rods, although they are the less sensitive (Haines, 1939; Feeney & Nagy, 1952; Garibaldi, 1967). This study demonstrated that 10% (w/v) of egg white (0·002% of conalbumin) is the optimal quantity to add to media without compromising the growth of pseudomonads, and the amount beyond which bacterial growth is influenced negatively. Tannic acid also significantly enhances FP production at a concentration of 5 μm.

The growth reduction or inhibition caused by conalbumin to several Gram-positive bacteria can be overcome by supplementing with sucrose (Feeney & Nagy, 1952). In fact, the preliminary results confirmed that bacterial growth was lower on PGS agar medium without or at very low sucrose concentrations. Moreover, the choice of growth factors, such as peptone, in a medium is very important for the efficient detection of fluorescence. The type of peptone, the composition and the concentration to be used was reported by Georgia & Poe (1932). Some peptones lack the necessary amount of the different constituents (magnesium, phosphate and sulphates) that are essential for pigment production, which must always be present, in addition to the organic constituents, for the production of FP in synthetic media by members of fluorescent Pseudomonas spp. (Georgia & Poe, 1931). In the media here, Bacto peptone and two different salts were used as sources of magnesium, phosphate and sulphates.

In general, in order to be assimilated by the bacteria, iron must be readily available in a medium; however, under aerobic conditions at a neutral pH, iron forms insoluble Fe (III) oxide hydrates that are not readily available. Under such conditions, bacteria start to produce several factors that are necessary to acquire iron, such as toxins that facilitate iron release from different sources and enzymes that destroy iron-binding proteins. To overcome this problem of iron accessibility in the pseudomonads, bacteria produce iron-chelating molecules called siderophores. These molecules solubilize ferric ions and transport these ions into bacterial cells via specific outer membrane transporters (Braun, 2003; Schalk, 2008; Lamont et al., 2009). The results presented here on the fluorescence detected on media with and without IBS and ferric iron supplements are in agreement with this report.

Some pseudomonads such as P. syringae pv. actinidiae, despite their ability to produce FP, do not do so on KB and CSGA media due to trace amounts of iron. Because the removal of iron metals by precipitation is a time-consuming and expensive practice, the incorporation of an iron-binding agent represents a rapid and economic solution to make the same medium a more efficient diagnostic tool for detecting FP production. The inefficiency of KB as a fluorescence detection medium for some pseudomonads was previously reported by Luisetti et al. (1972), although they did not study the role of iron availability as an inhibiting factor in siderophore production. In fact, they proposed CSGA medium for Pseudomonas syringae pv. persicae; however, the CSGA medium did not support FP production from P. s. pv. actinidiae (Scortichini et al., 2002). Species such as P. syringae pv. mendocina and P. corrugata do not excrete the FP on any of the media with or without an IBS supplement, probably because of the lack of genes that encode the enzyme involved in fluorescence expression (Budzikiewicz, 2004).

Among the IBS, use of the commercially available conalbumin is suggested, because the preparation of egg whites is a time-consuming task. The validation assay carried out through an intra-laboratory study confirmed the higher efficacy of the medium proposed in this study compared to that of commonly used media. In addition, the effectiveness of PGS agar supplemented with egg whites over time proved that the medium did not degrade and maintained its capacity even when stored at room temperature. The reduction of FP observed under UV over time occurred because of the development of an opaque coloration. Surprisingly, most of the pseudomonads that were fluorescent on KB medium lost their fluorescence after 36 h on PGS agar and became opaque, whereas species of Pseudomonas that were non-fluorescent on KB, like P. syringae pv. actinidiae, maintained their fluorescence for 72 h.

Acknowledgements

The authors would like to thank Dr Giorgio M. Balestra for providing all the strains of Pseudomonas syringae pv. actinidiae. Many thanks to Dr Mariagrazia Antonelli and Mrs Claudia Bartoli for their kind support in the intra-laboratory validation study.

Ancillary