FliI Role in Flagellar Assembly of Salmonella ΔfliI Mutant Strain Determines Motility and Biofilm Formation

Iram Liaqat1,*, Safdar Ali Mirza2, Sumera Sajjad3, Shaukat Ali1,*, Muhammad Faiz Qamar4 and Ikram Ul Haq5 1Department of Zoology, Govt. College University, Lahore-54000 2Department of Botany, Govt. College University, Lahore-54000 3Department of Zoology, Lahore College for Women University, Lahore 4Department of Pathobiology, College of Veterinary and Animal Sciences, Sub-campus Jhang 5Institute of Industrial Biotechnology, Govt. College University, Lahore-54000 Article Information Received 28 June 2019 Revised 30 July 2019 Accepted 18 August 2020 Available online 10 December 2020


INTRODUCTION
M otility and biofilm formation plays important role in bacterial pathogenesis hence making this aspect ideal to understand bacterial physiology. Almost every microbe including Pseudomonas aeruginosa, Vibrio cholerae, Salmonella and pathogenic E. coli, has ability to show motility but how this motility contribute to biofilm forming ability has not thoroughly studied (Chevance and Hughes, 2008).
Bacteria exhibit both swimming and surface swarming. In aqueous medium, microbes prefer swimming motility while on semi-solid surfaces, swarming, collective migration of bacteria occurs (Fraser and Hughes, 1999). For swarming motility, vegetative bacteria undergo a process of elongation and hyper flagellation which make O n l i n e F i r s t A r t i c l e secretion system (T3SS). This system contains six integral membrane proteins: FlhA, FlhB, FliO, FliP, FliQ, FliR (for Salmonellae and other species) at least. Among these flagellar proteins, FliI is the only established ATPase. It interacts with FliJ (no known function), and with a dimer of FliH (an inhibitor of FliI). ATP hydrolysis caused by FliI is important factor in gate-activation process. This suggested that FliI plays important role in energy provision to T3SS (Minamino et al., 2014). Besides swimming motility, S. typhimurium is among earliest serovars to show morphological differentiation of swarmer cells (Harshey et al., 1994). Kim and Surette (2005) studied swarming motility in S. typhimurium and linked it to evolutionarily conserved behaviour in Salmonella. Therefore, in this study, we have looked into both swimming and swarming motility of fliI complemented S. typhimurium to check if surface swarming exist in S. typhimurium. Furthermore, we investigated the importance of flagellar mediated biofilm formation in perspective to FliI complementation in SJW2702 (Δ fliI) strain. We constructed fliI complemented strains by overexpressing FliI using pTrc99A vector and showed that FliI deleted SJW2702 (Δ fliI)) strains are inefficient in energy coupling mechanism of flagellar type III protein export system making them aflagellated. To our knowledge this study is first of its type to demonstrate the role of fliI gene in flagellation and biofilm formation

Salmonella strains and culture conditions
Bacterial strains Salmonella enterica serovar Typhimurium SJW1103 and SJW2702 (ΔFliI) were used in this study. These were already available in lab obtained from Yamaguchi et al. (1986) and Kubori et al. (1992). The bacteria were routinely cultured in Luria-Bertani (LB) broth and agar at 37°C. When necessary, chloramphenicol (50 µgml -1 ) was supplemented to the growth medium.

Growth analysis
Culture samples were withdrawn at regular intervals to measure the optical density (OD 600 ). Replicate growth curve data from the SJW1103 (wild type), SJW2702 (Δ fliI) and fliI complemented strains were analyzed by drawing a logarithmic scale through the exponential-growth data points for each experiment (Riaz et al., 2018). Slope was used to calculate the specific growth rate constant.

Motility assay
Sixteen h old wild type and fliI Salmonella cultures were grown in motility medium supplemented with 0.1% glycerol and chloramphenicol (50 µgml -1 ). Swimming motility was observed by incubating plates for 5 h without being inverted. For swarming motility, a region extending ~1 cm into the colony was observed by phase contrast microscopy following Turner et al. (2010).

Fluorescent staining of flagella
Fluorescent staining of flagella was performed by following protocol by Turner et al. (2010). Briefly, swarm cells were collected, washed and centrifuged. Pellet was gently suspended in motility medium and thiol-reactive dye (Alexa Fluor 488; Invitrogen-Molecular Probes) was added. Cells were washed with motility medium and Image J was used to measure the lengths of the cell bodies and numbers of flagella (http://rsb.info.nih.gov/ij/). This information was used to measure the polymorphic transition in flagella following Calladine (1975).

Biofilm formation analysis
Biofilm assay was performed as described previously by Liaqat et al. (2016). This was done in two stages. In first stage, time kinetics for biofilm formation by all three strains was performed following Liaqat and Sakellaris (2012). Second stage of biofilm formation was performed using two assays. In test tube essay, LB medium supplemented with antibiotic was prepared and inoculated and incubated for 96 h. In air-liquid interface method,

O n l i n e F i r s t A r t i c l e
nutrient broth solution was inoculated and poured in petri plates. Coverslips were very cautiously placed aseptically followed by measurement of optical density (OD 595 ). Both test tubes and air liquid interface coverslip assays were performed two times for all Salmonella strains, and the averages and standard deviations were calculated for all repetitions of the experiment.

Statistical analysis
Statistical analyses were performed using student "t-test" for independent samples. All the experiments were performed three times. Data was analysed using Microsoft Excel and SPSS 18. The level of significance was P<0.05.

FliI cloning and SDS-PAGE analysis of Salmonella typhimurium
To verify the role of FliI, we constructed fliI complemented strains. Figure 1 shows 1.4kb band of interest on the agarose gel electrophorses using phusion PCR amplification. This fragment encoding fliI was subcloned into pET-28a (+) vector. Subsequent restriction digestion and sequencing verified the recombinant plasmid. The double enzyme digestion of the recombinant plasmids demonstrated successful ligation into the vectors. Afterwards, sequencing confirmed that the sequence of cloned fragment to be accurate (data not shown).
We introduced fliI overproduction plasmid into Salmonella SJW2702 (Δ fliI). SDS-PAGE anlaysis of each salmonella strain revealed that the amounts of overexpressed fliI was almost similar in both SWJ2102 (wild type) and fliI complemented strains, as seen in sodium dodecyl sulfate gels (Fig. 2, 2 nd and 3 rd lane). There was no apparent FliI bands in deletion mutants of Salmonella SJW2702 (Δ fliI) (Fig. 2, 1 st lane), fliI bands were seen in wild type even in the absence of IPTG. In the presence of 2mM IPTG, fliI bands was the major ones in the whole-cell extracts (Fig. 2, 4 th lane).

Growth and motility assays
There was no calculated difference in growth rate of bacteria as was observed by their growth curve and specific growth rate calculations (Fig. 3A, B). Comparison of swimming and swarming motility assays among three strains showed that SJW1102 (wild-type) and fliI complemented Salmonella strains produced a large swim ring while the fliI deletion mutant Salmonella SJW2702 stayed at point of inoculation showing no motility (Data not shown). Phase contrast microscopy of fliI complemented cells showed that all cells had flagella (Fig. 4A). Following Calladine (1975), polymorphic transitions were observed in flagellated strains. Most of the flagella in our study were semicoiled in nature (Fig. 4B). To understand the role of fliI in swarming motility, we grew bacteria on swarm agar plates containing LB and 0.35% agar. SJW1102 and fliI complemented strains displayed motility on media. However, SJW2702 (ΔfliI) failed to exhibit any swarming motility on 0.35% agar. Fig. 3. A, S. typhimurium growth curves. Three strains of Salmonella including wild type, S. typhimurium complemented fliI and SJW2702 (ΔfliI) were grown in Luria-Bertani broth at 37°C with aeration. Bacterial growth was determined by OD 600 . Data were obtained from the average of three independent experiments; B, specific growth rate of three strains.

Biofilm formation study
The biofilm-formation by all three strains of Salmonella was quantified using crystal violet staining method. Results of biofilm time kinetic indicated that all tested Salmonella isolates produced strong biofilm after 96 h. Afterwards, a decline in biofilm formation was observed (Fig. 5A). Once biofilm-forming capacity of different Salmonella isolates was assessed, we investigated further the difference in biofilm forming capacity of all three isolates using test tubes and liquid interface coverslip assays. We were interested to see whether fliI complemented strain has good biofilm forming capacity similar to swimming and swarming motility. As expected, strong biofilm formation by fliI complemented strain was observed compared to both wild type and fliI deleted strains strain (Fig. 5B, C) using both assays.

DISCUSSION
Flagellar apparatus has been observed to play crucial role in pathogenesis of a great diversity of intestinal pathogens. In this regard, the flagellar assembly of S. typhimurium is an interesting aspect to understand bacterial adherence mechanism and biofilm study. There is not much data about translocation of flagellar proteins from the cytosol to the distal end. Among several proteins studied so far, FliI is especially interesting because of its identical nature to catalytic β subunit of the F 1 -ATPase and homology to various proteins in T3SS. Majority of bacteria including both Gram positive and Gram negative have FliI playing ATPase role in type three secretory system (T3SS). In this study, we have analyzed FliI role via complementation in flagellum-mediated rotation for swimming, swarming motility and biofilm formation.
We observed that FliI complemented strains have no difference in growth compared to motility and biofilm formation. Optimum swimming and swarming motility independent of planktonic growth observed in wild type and FliI complemented strains compared to SJW2702 (Δ fliI) is related to fact that fliI deletion lead to lack of energy for flagellar export. In a previous study, novel motility regulators were screened by genomic analysis and 130 mutations were found to be important to influence motility in S. typhimurium genome (Bogomolnaya et al., 2014). In fact, two energy sources used by flagellar export system include ATP and proton motive force (PMF). FliI forms a homo-hexamer and is the only ATPase of the export system. Although FliI makes export gate highly highly efficient, however, its role is still unclear because of limited information about ATPase mechanistic nature (Minamino et al., 2014).

O n l i n e F i r s t A r t i c l e
I. Liaqat et al. It has been documented in several previous studies that either complete flagella or parts of it could promote bacterial adhesion and binding to the host's surfaces thus enhancing virulence. Biofilm formation is an adaptation by different bacterial species to enhance survival and pathogenesis. We observed biofilm formation by all Salmonella isolates in this study. This might be due to greater glycocalyx production at that stage as reported by MacFarlane et al. (2007). However, Salmonella strain lacking fliI gene exhibited almost one fold decrease in biofilm formation compared to wild type and fliI complemented strain. Wood et al. (2006) demonstrated that flagella are important both in biofilm initiation and development. Likewise, Olsen et al. (2013) reported that serovar-specific differences are important in determining in the flagellar involvement as well as chemotaxis genes in attachment and invasion of Salmonella to the host Following time kinetics of biofilm formation, its quantification was performed using test tubes and liquid interface coverslips assays. Significantly decreased biofilm in fliI deficient strains observed in this study might be due to lack of flagella. Non-motile flagellar mutants might have decreased initial surface attachment hence showed poor biofilm formation (Liaqat et al., 2016). Additionally other factors including fimbriae, pili, curli may also contribute to decreased biofilm formation in non flagellar strains (Reisner et al., 2006;Lemon et al., 2007;MacFarlane and Dillon, 2007;Kim et al., 2008;Liaqat and Sakellaris, 2012). Importantly, higher biofilm formation observed by fliI complemented strain even compared to wild type. This might be due to the fact that FliI overexpression essentially means more energy production for flagellar export leading to enhanced flagellation hence initiating formation of biofilm. Previous studies by Lemon et al. (2007) and Gorski et al. (2009) are consistent with our finding that without flagella or flagella motility, biofilm formation was significant reduced.
The present study suggested that motility is stringently controlled by an organized flagellar assembly. We concluded that FliI is essential for flagellation, motility and biofilm formation in Salmonella strains. We are the first one to report that without FliI, motility is essentially lost leading to significantly reduced biofilm formation in S. typhimurium. More detailed studies on this aspect will lead to better understanding of various mechanisms involved in Salmonella motility at molecular level. Among several questions about the role of flagella, one question which we are interested to address was the role of flagellation in attenuation of strain pathogenesis as was observed by Yang et al. (2012). Efforts will be made in our lab to study fliI complemented role through in vivo trails.

O n l i n e F i r s t A r t i c l e
FliI Role in Bacterial Flagellation and Biofilm Formation