The Study of Microbial Communities of Rudkhan Castle

Abstract Rudkhan Castle, one of the most valuable monuments in Iran, belongs to the Sasanian Empire, and nearly all of the parts are made of bricks. The castle had been exposed to physicochemical and biological factors over the years. Locating in a humid environment and possessing porous surface has made it appropriate substrata for microorganisms’ colonization. This study is an attempt to identify the major microorganisms involved in biodeterioration of Rudkhan Castle. The samples were taken from brick surfaces and were sub-cultured onto culture media. Scanning electron microscope (SEM), stereomicroscope and Periodic Acid Schiff (PAS) staining were also employed to illustrate the growth and pattern of penetration into the substratum. Energy dispersion X-ray analysis (EDX) and X-ray diffraction (XRD) were used to show alteration of elements of deteriorated brick samples, which are caused by microorganisms’ activities. First, Samples were identified by morphological characteristics. Next, some of the isolates were identified according to molecular methods, polymerase chain reaction (PCR). The first step to protect the cultural heritage is identification of deteriorating agents.


Introduction
A huge number of monuments around the world are suffering from deterioration and degradation. In recent decades, a diverse microbial community isolated from archaeological sites has been reported (Qiang et al. 2016). Growth of living organisms on monuments can cause alterations and physicochemical damages, which is known as biodeterioration. Colonization of bacteria, lichens, mosses, fungi, and algae depends on environmental factors such as humidity, light intensity, air pollution, climatic conditions, porosity and mineral components, and the bio receptivity of the substrata (Crispim et al. 2003;Pandey et al. 2011). The exact interactions between environmental factors and microorganisms, which affect monuments, are not well understood (Bhavani et al. 2013). Phototrophs including cyanobacteria and microalgae are the pioneer colonizers of rock, brick and marble surfaces, and have been suggested to be of greater ecological importance since they produce their own nutritive requirements (Albertano et al. 2000). They provide an area for the growth of microbial biofilms, including bacteria, fungi and lichen. These organisms can cause degradation of stone by the production of aggressive acid or alkaline metabolites and surfactants. Furthermore, they damage the stone via physical penetration of their cells into the substrata (Gaylarde and Gaylarde 2004). Colonization and deterioration of stone monuments by biofilms causes extensive, aesthetic, physical, and chemical damages. The effects of these damages are manifested as discoloration, pitting and etching.
Rudkhan Castle, one of the most valuable monuments in Gilan province, Iran, which belongs to the Sasanian Empire (224-651 AD) (Fig. 1). The main materials of the castle consist of brick with mortar of sarooj 1 and lime. The castle is located at 37 13 0 15 00 N latitude, 49 18 0 30 00 E longitudes and altitudes of 665-715 m above sea level on the top of the mountain. It includes 2.6 hectares of wetlands. In the whole year, the mean annual temperature is 15.7 C and the mean annual precipitation is 1275 mm. The average relative humidity is 90% and never drops below 70% ( www.Fouman-city.ir 2017). Locating in a humid environment and possessing porous surface, the monument has provided perfect substrata for colonization of microorganisms. There is little information on microbial community of Rhudkhan castle and yet no study has been done on its biodeterioration agents.
In this study, we aim to identify microflora, colonizing different parts of the castle using sequencing of 16 s rDNA for cyanobacteria, 18S rDNA for eukaryotic microalgae and ITS sequences for lichen and show the possibility of damages caused by these microorganisms. Scanning electron microscope (SEM), X-ray diffraction (XRD) and Energy dispersion X-ray analysis (EDX) and Periodic Acid Schiff (PAS) staining were used to show the presence of microorganism on the surface and physicochemical effects of them on the surface.

Sampling and culturing
Samples were taken in September 2015 from 28 different areas of vertical and horizontal brick surfaces of the castle where there were signs of deterioration (Fig. 2). Black spots, discoloration, pitting, powdery substances were signs of deterioration on bricks that were considered while sampling. All of the samples were coded and transferred to the laboratory. Algae and cyanobacteria were isolated from the samples by needle technique and adhesive tape stripes and then were sub-cultured onto blue-green algae medium number 11 (BG-11) and Bischof & Bold medium (BBM). Fungi were also isolated by needle and adhesive tape stripes and were sub-cultured onto potato dextrose agar (PDA) and Sabouraud dextrose agar (SDA). Lichen samples were collected using sterile scalpel then were glowed on cardboard. Those lichens that firmly clung onto the substratum were scratched and collected in microtubes for molecular identification. Also broken pieces of bricks that they had fallen on the ground were taken to the laboratory in order to be used as samples for SEM and XRD. Phototrophic isolates were incubated at 25 C in a light:dark cycle 8:16 h, incubation time was different depending on the type of isolates from 14 days to a month (Anderson 2005), and fungal isolates were kept at 25 C and incubation time was 7 days.
Microscopic structures of fungi, particularly characteristics of spores, conidia, and budding features were observed using an optical microscope (Nikon YS100, Chiyoda, Japan) and isolates were identified according to related references (Ellis et al. 2007;Watanabe 2002). Morphological identification of lichen isolates was performed in the herbarium of Research Institute of Forest and Rangeland of Iran. Phototrophic cyanobacteria and microalgae were observed by optical microscope and were morphologically identified according to valid references (Prescott 1970;Wehr and Sheath 2003).

Scanning electron microscope (SEM)
SEM and stereomicroscope (AIS2100 Seron, Uiwang, Korea) were used to evaluate the presence of the microorganisms and to observe the morphological structures of biological growth on the samples. For these reasons, tiny fragments of brick were fixed using 4% glutaraldehyde for 3 h and were subsequently washed in 1% phosphate buffer. Then, different series of ethanol concentrations (30, 50, 70, 80, 90, and 100%) were used to dehydrate the samples (Herrera and Videla 2009). Finally, the samples were dried and coated (Emitech SC70R0 sputter coater, UK) with gold for 90 s in vacuum conditions.

X-ray diffraction (XRD) and energy dispersion X-ray analysis (EDX)
XRD (X'Pert PRO MPD PANalytical, Nottingham, UK) and EDX (AIS2100 Seron, Uiwang, Korea) were used to compare compounds of those samples which were covered by microorganisms and the control one.

Molecular methods
Identification of some isolates was difficult through the morphological methods that were used; they were therefore identified via molecular techniques. DNA extraction using QIAamp DNA Mini Kit (250) (Qiagen, US), was performed for lichen according to manufacturer structure. DNA extraction for cyanobacteria and microalgae was performed by liquid nitrogen at the first stage; In such a way that 0.15 g of not dried sample was powdered by liquid nitrogen. Next 0.15 ll of extraction buffer containing 100 ml of Tris 1 M (pH 8), 97.1 ml of EDTA, 0.5 M (pH 8), 14.61 g of NaCl, 25 g of SDS and the distilled water in a final volume of 1 l of solution, was added after incubating at 65 C and was mixed for 30-60 s. 200 ll of potassium acetate was added to the mixture. The mixture was mixed for 5 min and was put on the dry ice for 10 min. Next 500 ll of phenol:chloroform:isoamyl alcohol (25:24:1) solution was added. The mixture was centrifuged for 15 min in 5000 rpm. The upper solution was transferred to another tube and was mixed by cold isopropanol with equal amount. In this stage the mass of DNA was visible. The supernatant was removed and sediment was washed twice by ethanol 70%. After removing ethanol 70%, the tube was put under laminar flow hood for 30 min until complete evaporation of ethanol and water. Finally the mass of DNA was solved in 50 ll of sterile DW water (Eland et al. 2012). Polymerase chain reaction (PCR) process was performed by PeqSTAR hpl Gradient (Isogen, Netherlands) device. In this study, the primers of ITS1F (CTTGGTCATTTA GAGGAAGTAA) and ITS4 (TCCTCCGCTTATTGAT ATGC) were used to amplify the DNA lichen and the size of amplified regions was 800-1000 bp. The master mix buffer contained 2.5 ll PCR buffer, 0.75 ll MgCl 2 , 0.5 ll dNTP, 0.5 ll ITS1F, 0.5 ll ITS4, and 0.5 ll taq DNA polymerase (Brown 2010;Mohammadi and Maghboli-Balasjin 2014;Primrose et al. 2001).
The quality of the PCR products was tested by gel electrophoresis. Then, The PCR products were sent to Sequetech Company to be sequenced by Sanger method.

Thin section
Thin sections were made from small slabs of brick samples which were then glued to a glass slide. Then, they were ground to a specified thickness of 30 microns. At this thickness, most minerals become more or less transparent and could then be studied by a light microscope. Brick samples taken from Rudkhan Castle were used to demonstrate the formation of pitting and other eroded structures through the action of fungal hyphae and their fruiting bodies as well as algal and bacterial cells.

Periodic acid Schiff (PAS) staining
The PAS staining is a histochemical reaction in that the periodic acid oxidizes the carbon-carbon bonds. In fact, this  is an appropriate method to stain carbohydrate-rich microbial biofilms. This stain reacts with organic substances such as protein-carbohydrate, starch, glycogen, etc. This produces aldehyde groups, which can then condense with Schiff's reagent, forming a bright red coloration. The periodic acid schiff reaction is usually used to stain brick biofilms. The mycobiont penetration of lichen or free-living fungi can be visualized on or inside brick (Mohammadi and Krumbein 2008). The PAS staining of the thin sections of the brick biofilms was performed according to Whitlatch and Johnson (1974).
PDA and SDA media were used for morphological identification of fungi. On the basis of micro and macroscopic observations, Alternaria sp., Aspergillus sp., Ulocladium sp., Fusarium sp., Rhizopussp., and Penicillium sp., were identified (Ellis et al. 2007;Watanabe 2002). Those fungi which did not show any particular feature are reported as sterile fungi (Figure 4(a-f)). In this study, 10 lichen samples were obtained from the central part of Rudkhan Castle. Identified species belong to Verrucaria sp., Caloplaca sp., Dirina sp., Dermatocarpon sp., and Gyalecta sp.
In the earlier observations, the stereomicroscope (Nikon SMZ18, Chiyoda, Japan) showed a variety of biological structures such as cyanobacteria, algae, lichen, and mosses. On the surface of the brick, different alterations were observed such as salt crystallization, powdery layers of bio-mineralization, patinas and pitting ( Figure 5). Furthermore, SEM micrographs emphasized the presence of bacteria, cyanobacteria, algae, lichen, and fungi on the brick samples, and also the resulting pattern of bio-pitting, crystallization, and etching from microbial activities ( Figure 6).
Metabolites produced by microorganisms can cause alterations of the substratum. Likewise, a high concentration of iron on brick samples covered by biofilms could be an indication of the presence of microorganisms and their roles to chelate the cations. These data were supported by EDX profiles (Figure 7). In addition, XRD spectra showed the increase of calcite crystals as a result of a dissolution process, although the amounts of quartz, albite, and orthoclase were changed (Figure 8) Phylogenetic trees were constructed based on the neighbor-joining method using Poisson model with 1000 Bootstrap replications on MEGA 5.0 software after ClustalW alignment. Each isolate is showed by its accession number, Nostoc pruniforme MK144664, Nodosilinea epilithica MK144663, Gyalecta jenensis MK773874, Verrucaria    (Figures 9-11).
Thin sections of bricks were stained with acid periodic schiff reagent. The presence of lichen thalli on the brick surface and bio-mineralization of the substratum were well observed. Furthermore, the penetration of fungal hyphae to a depth of 3 mm of the thin sections and several deep cracks were detected. These results indicate towards the biodeterioration of Rudkhan Castle bricks through the presence and activity of microorganisms. Images of the thin brick sections before and after staining are shown in Figures 12 and 13.

Discussion
Biodeterioration of monuments depends on different factors such as air pollution, climate, humidity, light and temperature and features of the substrata. According to the studies, Figure 11. Phylogenetic tree depicting the relationships among lichen samples using the ITS. Bootstrap values are given above 50%. such microorganisms with close similarity have been isolated from monuments all over the world due to the ability of colonization on the stone surfaces. It seems that soil of the region is one of the most important origins of monument microorganisms (Bhavani et al. 2013).
Microbial identification based on the culture-dependent methods, is not always effective, because some microorganisms can change their morphological and physiological characteristics in the different conditions. Therefore, using molecular methods to identify microorganisms should be carried out (Gonz alez and Saiz-Jim enez 2005).
In this study, 13 isolates were detected based on the sequencing method. Sequencing of small subunits (SSU) 16S ribosomal RNA gene was successfully performed on cyanobacteria. Also carried out the sequencing of 18S ribosomal RNA gene on microalgae and ITS region for lichens and fungi carried out.
In earlier studies, phototrophs such as Phormidium, Leptolyngbya, Nostoc, Scytonema, Klebsormidium were isolated by Kumari, et al in 2008 from brick and stone monuments in India which have humid and warm climate (Kumaril and Prads 2008) and majority of these phototrophs were also found in Rudkhan castle which shows similar pattern in both region. Filamentous microalgae and cyanobacteria were also found in this study such as Leptolyngbya, Nostoc, Scytonem, and Klebsormidium.
Tallus fungi such as Alternaria sp., Aspergillus sp., Fusarium sp., Rhizopus sp., Ulocladium sp., and Penicillium sp. can cause damages by penetrating to the surfaces as it is shown in scanning electron microscopy images. Moreover, Pandey et al. (2011), in their study mentioned that fungi such as Alternaria sp. can cause discoloration as well as mechanical destruction of stone and some species of genus Aspergillus sp., cause damages by secretion of organic acids.
Moreover, four lichen species were reported as existing in Hassan Tower in Morocco including Caloplaca vitellinula, Caloplaca flavescens, Roccella phycopsis, Xanthoria calcicola (Nattah et al. 2012). Based on this study lichen can cause damages by penetration of their fungi hyphe to the substratum and also by secreting oxalic acid carbon. As a matter of fact, lichen and biofilms generally tend to grow on the upper surfaces of the brick which are exposed much more to the humidity. Considering the nature of bricks and their porous structures, this kind of biodeterioration has a significant role in the destruction of Rudkhan Castle and is speeding up the amount of deterioration. With the help of SEM, PAS staining, (Charola et al. 2009), concluded that Rudkhan Castle bricks have a significant cover of lichens, such as the Parmelia, Caloplaca, Pyxine, Dirinaria, and Lepraria genus (Charola et al. 2009).
It can be concluded that the microorganisms have a crucial role in the biodeterioration of Rudkhan Castle. The first step to protect cultural heritage must be the detection and identification of deteriorating agents and consequently, the best decision should be selected for the conservation of this marvelous monument.