Limestone samples were collected from the study site (El-Amir Muhammad’s Minaret, Akhmim, Sohag, Egypt) by non-destructive methods.

Nikon polarizing microscope (JEOL JSM5500LV) was used in the petrographic study of limestone samples.

XRD Unit, Faculty of Science, Assuit University, Model PW 1710 control unit, Philips, 2θ from 4 to 60, Anode Material Cu, 40 K.V, 30 M.A.

Identification of the chemical composition of limestone samples was carried out using (Axios Advanced, Sequential wd_ XRF Spectrometer, PANalytical 2005) in the Analysis Unit of the National Center for Building Materials Research in Cairo.

SEM was used to study and understand the fine structure, decomposition, and different properties of the limestone under study. It was carried out in the Central Lab, South Valley Univ. using JEOLJSM-5500 LV SEM (JEOL, Japan).

The settle plate method [22] was used to estimate the airborne spores in of El-Amir Muhammad’s minaret. Nutrient agar and Czapek’s (CZ) agar media were used for isolation of bacteria and fungi, respectively. The plates were exposed for five minutes. Nutrient agar plates were incubated at 37˚C for 72 h while CZ plates were incubated at 28˚C for 7 days. The developed colonies were counted in plates and the average number of colonies per three plates was determined.

Dry cotton swabs were rubbed on the surface of the deteriorated parts of the building over an area of 4 cm², under aseptic conditions, stored at 4˚C until used for inoculation as mentioned previously.

The bacterial isolates were tentatively identified on the basis of classification schemes published in Bergey’s Manual of Systematic Bacteriology [23].

The Bacterial isolates were cultured in sterile test tubes containing 10 ml of nutrient broth medium [24]. The cultures was incubated at 28˚C for 48 hours, then sent to the molecular Biology Research Unit, Assiut University for DNA extraction using Patho-gene-spin DNA/RNA extraction kit provided by Intron Biotechnology Company, Korea. From each sample the DNA was sent to SolGent Company, Daejeon South Korea for polymerase chain reaction (PCR) and gene sequencing. PCR was performed using two universal primers where 27F (5'-AGAGTTTGATCCTGGCTCAG-3') and 1492R (5'-GGTTACCTTGTTACGACTT-3') were used. The purified PCR products (amplicons) were reconfirmed using a size nucleotide marker (100 base pairs) by electrophoreses on 1% agarose gel. The amplicons were sequenced with the incorporation of dideoxynucleotides (dd NTPs) in the reaction mixture. Bacterial amplicons were sequenced in the sense and antisense directions using 27F and 1492R primers [25]. Sequences were further analyzed using Basic Local Alignment Search Tool (BLAST) from the National Center of Biotechnology Information (NCBI) website. Phylogenetic analysis of sequences was done with the help of MegAlign (DNA Star) software version 5.05. Sequences were aligned with others retrieved from Gen Bank using ClustalX [26] and optimized manually. The positions where one or more species contained a length mutation and ambiguously aligned regions were not included in the subsequent phylogenetic analysis. Maximum parsimony and maximum likelihood analyses were made in PAUP 4 [27]. Maximum-parsimony (MP) trees were obtained by 100 random addition heuristic search replicates, and 1000 bootstrap replicates were performed employing 5 random addition heuristic searches. Maximum-likelihood (ML) analysis [28] was performed using heuristic searches with the random stepwise addition of 100 replicates and tree bisection-reconnection (TBR) rearrangements. The optimal model of nucleotide substitution for the ML analyses was determined using hierarchical likelihood ratio tests (hLRTs) Model test 3.7 [29]. The model selected as the best fit for 16s rDNA dataset was TrN. Phylogenetic trees were visualized using Njplot [30] and edited in Adobe Illustrator CS6.

Identification of fungal isolates was performed according to Raper and Thom (1949) [31]; Gilman (1957) [32]; Domsch (1980) [33].

Bacterial and fungal isolates were tested for calcium carbonate dissolution by growing colonies on Deveze-Bruni (DB) medium, consists of (g∙L⁻¹): glucose, 5 g; yeast extract, 1 g; peptone, 1 g; K₂HPO₄, 0.5 g; MgSO₄, 0.01 g, NaCl, 5 g; NH₄(SO₄)₂, 0.05 g; MgCl₂, CaCO₃ 5 g and 1.5% agar), bacteria were incubated at 37˚C for 7 days, fungi were incubated at 8˚C for 14 days [34]. Bacteria and fungi that dissolve CaCO₃ can be distinguished due to the apparent halo of a clear zone around the colony.

The microcide solution was prepared by dissolving sodium azide ethyl alcohol 95% to give concentrations ranging from 50 up to 150 µg/l (ppm). The most dominant bacterial and fungal isolates were treated by different concentrations (25, 50, 100 and 150) ppm of sodium azide using the agar well diffusion method [35]. Control without microcide was done using ethyl alcohol. Inhibition zones were measured to determine the minimum inhibition concentration

In the present case, limestone in general is quite homogeneous in its chemical characteristics, being dominated by CaCO₃, but is highly variable in terms of physical characteristics such as hardness, fossil content and porosity [36] [37]. Limestone, which is often used in stone construction, where the effects of sulfate attack appear very quickly, especially in cases of stone damage, the formation of gypsum crusts, results in stone cracking [38] (Figure 2(b) and Figure 2(c)). Humidity increases the chances of damage due to wet and dry thawing and freezing. Where the soluble salts move in and out of the components of the porous stones. Seasonal changes in soil and temperature also control, which leads to the appearance of salt flowers that are soluble (Figure 2(a)). Weathering is also related to mechanical strength, water absorption and permeability of the stone and treatment work must include the estimation and improvement of these properties. The pore space in the weathering zone of limestone becomes clogged when the process of recrystallization of the salts takes place as a result of damage to its components and cause discoloration of limestone [39] (Figure 2(b)). Water has an important role in the damage of limestone because it is highly polar [40]. It also helps in the rate of deterioration of limestone, the high content of clay minerals, which are expandable minerals. The extent of the ability of limestone to absorb moisture is expressed in estimating the clay minerals content, as water retention is a result of the presence of clay minerals in the pores. The presence of clay minerals in the formation of limestone in addition to the characteristics of limestone such as its high porosity causes severe damage to limestone caused separation and loss of some stone blocks (Figure 2(c) and Figure 2(d)). Many cracks were observed due to swelling and contraction of the components of the clay minerals due to changes in relative humidity, and the expansion and contraction of the degraded limestone components as a result of the changing temperature cycles. The degree and depth of damage also indicates that the components of the minaret have been damaged by water due to previous treatments, as they appear as visible white sediments caused by microbiological damage on the stone surface Figure 2(e) and Figure 2(f).

Optical microscopy revealed some details such as porosity, fine cracks and grain morphology, the grains formed a fragmented matrix of heterogeneous, non-solid powders, with a large amount of fine cracks. The study showed that the limestone grains contain skeletal and fossil parts of fossils of the foraminifera, especially the nemolite, which consist of calcium carbonate CaCO₃ and dolomite, were found to fill the cracks and dissolve with some traces of gypsum CaSO₄∙2H₂O. Note calcite sparite in the inner spaces of the fossils, and the main mineral is micritic calcite as shown in Figure 3.

XRD results showed that calcite (CaCO₃) is the main component of limestone samples, and the presence of CuFeNaS₂ compound is an indicator of the presence of gypsum and halite with gypsum (CaSO₄∙2H₂O) (Figure 4).

Unexpectedly, XRD analyzes failed to determine the presence of aluminum silicate or clay minerals, as this was inconsistent with the results obtained from ESEM-EDS analyzes, where the latter was able to detect aluminum silicate within the polished samples and this may be due to the presence of compounds Clay

minerals in amounts below the detection level of XRD analyzes in the analyzed samples (about 5%).

Scanning electron microscope (SEM) in Figure 5 clearly show that there are details in the morphology of the stone surface, especially the appearance of small and round white particles, and the preliminary analysis showed the presence of calcium (Ca), carbon (C) and oxygen (O) in most of the samples, which may indicate that these particles calcium carbonate. The grain distribution appeared inhomogeneous as silicon (Si) and aluminum (Al) Figure 6, which indicates the presence of aluminum silicate minerals (clay minerals), where they are found in the thin regions between the white particles as revealed by SEM images. The presence of fossil mollusks distributed throughout the matrix, and this indicates the origins of the limestone under study.

XRF study that Aluminum oxide Al₂O₃ shows an average percentage of 0.8102% which could be indicate to presence of clay minerals, and magnesium oxide MgO shows an average percentage of 0.5043% and ferric oxide Fe₂O₃ show an average percentage of 1.1726% and SO₃ shows an average percentage of 0.5672% which indicates to the presence of gypsum where Gypsum is formed when calcium carbonates CaCO₃ react with sulfur dioxide SO₂ Table 1.

Based on morphological, physicochemical and physiological characterization (Table 2), isolated bacteria were tentatively identified as Bacillus cereus, Bacillus subtilis, Micrococcus ruseus, Micrococcus luteus, Staphylococcus aureus, Streptomyces sp. Pseudomonas oryzihabitans and Serratia marcescens.

The 16S rDNA dataset included 20 sequences: 10 belong to the family Bacillaceae, 5 Pseudomonadaceae and 5 Yersiniaceae. The maximum parsimony dataset consisted of a total of 350 characters, of which 257 were constant, 9 variables and parsimony-uninformative and 84 were counted as parsimony-informative. Maximum parsimony produced 15 most parsimonious trees all had a length of 135 steps, a consistency index of 0.8889, a retention index of 0.9716, and a rescaled consistency index of 0.8637 and of which one is shown in Figure 7. Maximum likelihood analysis yielded a single most likely tree (–ln likelihood = 1135.93). Bayesian analysis yielded two trees similar in topology to the ML phylogenetic tree shown in Figure 7. The four bacterial strains nested within the genera: Bacillus (two species), Pseudomonas and Serratia. The bacterial strain

+ve = Positive, −ve = Negative, ND = Not Detected, O = Oxidative, F = Fermentative.

RM1 was grouped with Bacillusproteolyticus and B. albus in all the phylogenetic analyses perfor/med with high statistical support (93/89/100 for MP/ML/BYPP, respectively), while the bacterial strain RM2 was grouped with Bacillus subtilis and B. pizizenii with high statistical support (91/97/100 for MP/ML/BYPP, respectively). The third strain RM3 was grouped with Pseudomonas sp. (KP720613)

and P. oleovorans with high statistical support (93/89/100 for MP/ML/BYPP, respectively) and the fourth strain RM was grouped with several strains of Serratia marcescens (Figure 7).

Six airborne bacterial genera: Bacillus, Staphylococcus, Micrococcus, Streptomyces Pseudomonas and Serratia were recovered from out and indoor of El-Amir Muhammad’s minaret. The data in Table 3 show what Bacillus cereus OK447647 and B. subtilis OK447648 were the most common and the most frequent as well comprising (53.9%; 43.3% and 30.5%; 33.9%) of total counts from outdoor sand indoors, respectively. Gram negative bacteria were identified as Pseudomonas oryzihabitans OK447649 and Serratia marcescens OK447650 comprising (0.4%; 0.3% and 0.9%; 0.5%) of total counts from outdoor and indoor bacterial isolates, respectively.

Data in Figure 8(a) also reveal that Bacillus cereus was dominant, it was recorded from 100% of samples collected from indoor and outdoor aerosols of El-Amir Muhammad’s minaret, followed by Staphylococcus aureus and Streptomyces (60%; 100% and 60%; 60%), Serratia marcescens OK447650 and Pseudomonas oryzihabitans OK447649 were less frequent, in outdoors and indoors, respectively.

These results agree with the studies presented by Carlo et al. (2016) [41] who isolated spore forming bacteria: Bacillus thuringiensis and B. weihenstephanensis from the Saints Cave environment. On the other hand^, Awad (2007) [42] reported that Micrococci and Bacillus isolated from aerosols in a four-storey flourmill building located in Giza, Egypt were dominant where Gram-negative bacteria were found in low numbers.

The concentration and variability of the indoor airborne microbes can be affected by several factors, such as the infiltration of the outdoor air, the human presence and activities which give a good reason for the recognized increase of outdoor microbial counts compared to those of indoor counts of El-Amir Muhammad’s minaret. These findings come in agreement with Katsivela et al. 2021 and Stelzenbach 2002 [43]. On the other hand, Rajendran and Prasad (2012) [44] concluded that the numbers outdoor aeroflora higher compared to the indoor aeroflora of Vishnu temple in India.

B. cereus and B. subtilis were the most dominant among airborne bacteria recovered from El-Amir Muhammad’s minaret, where Gram-negative bacteria were less frequent, these results agree with the studies presented by Carlo et al. (2016) who isolated spore forming bacteria: Bacillus thuringiensis and B. weihenstephanensis from the Saints Cave environment. On the other hand^, Awad (2007) reported that Micrococci and Bacillus isolated from aerosols in a four-storey flourmill building located in Giza, Egypt were dominant where Gram-negative bacteria were found in low numbers.

Data in Table 3 and Figure 8(b) show that eleven airborne fungal species belonging to five genera were recovered from out and indoor of El-Amir Muhammad’s minaret. Aspergillus was the most prevalent genus represented by six species of which Aspergillusflavus was dominant it was recovered from 100% of samples comprising (34.2% and 23.8%) of total fungi from outdoors and indoors, respectively. Penicilliumchrysogenum was of moderate occurrence comprising (26.3% and 14.3%) of total fungi from outdoors and indoors, respectively, followed by Cladosporiumcladosporoids representing (10.5% and 22.2%) and was recorded from (100% and 80%)% of total fungi from outdoors and indoors, respectively. Lower frequencies were revealed by other genera as Alternariaand Syncephalustrum.

Among airborne fungi isolated from outdoor and indoor of El-Amir Muhammad’s minaret, Aspergillus flavus, A. niger and Penicillium chrysogenum were the most frequent. The presence of Alternaria alternate and Cladosporium cladosporoids was pointed out. Similar results were reported by (Carlo et al. 2016; Abdel Hameed 2009 [45]; Gillum and Levetin 2008 [46]. Airborne spores and cells may be carried by the wind or by human activities or deposited onto the wall surfaces by gravitational settling [47]. Most microorganisms are able to successfully grow on stone surfaces covered with dust, animal remains, air contaminants and secretion or finger-marks, creating invisible layer of biofilm (Rajendran and Nisy 2012).

According to data in Table 3 and Figure 9(a), eight species belonging to six bacterial genera were recovered from samples collected from surfaces of deteriorated limestone of outdoors and indoors of El-Amir Muhammad’s minaret. The genus Bacillus showed maximum frequency, it was represented by two species: B. cereus and B. subtilis comprising (54.5%; 42.9% and 24.2%; 32.1%) of total bacteria recovered from outdoors and indoors, respectively. Additionally, Bacilluscereus OK447647 was recorded all isolated samples. Micrococcus, Staphylococcus, Streptomyces, Serratia and Pseudomonas revealed lower counts comprising percentages ranging between (10.6% and 1.2%) of bacterial total counts. Comparable results were reported by Jroundi et al. (2020) [48], who

reported that Bacillus was detected in all samples of stones collected from Maya archeological site of Copan, Honduras. On the other hand, Banciu (2013) [49] stated that Strains of Bacillus spp. are some of most oftenly found bacteria identified on surface as well as inside the stone artifacts. Furthermore, Micrococcus, Staphylococcus, were recovered from pre-historic rock-paints of Kabra-pahad, India [50].

Data in Table 3 and Figure 9(b) depict that twelve species belonging to seven fungal genera were recovered from outdoor limestone of which Aspergillusniger was the most dominant, comprising the highest frequencies (36.7% and 47.3%) of total fungi from outdoors and indoors, respectively. A. niger and A. flavus were also reported from (77.8% and 88.9%) of samples collected from outdoor and indoor limestone, respectively. Lower frequency (6.3%) was comprised by Penicilliumcorylophyllum which was recovered from outdoor samples only of the total count and was recorded from 11.1% of samples. Dematiatiaceous fungi were represented by four genera (Alternaria alternate, Cladosporiumcladosporoids, CurvularialanataandUlocladium charatum) comprising percentages of 1.3% up to 11.5% of the total fungal counts. These results corroborate with that of Biswas (2013) who isolated eighteen fungal species from Kabra-pahad rocks in India among which Aspergillus group were the most dominant. [51] also illustrated that within the Aspergillus genus, the most abundant species was Aspergillus fumigatus Fresen., representing up to 28% of the outdoor fungi count, while Penicillium purpurogenum Stoll was dominant within its genus, with a maximum representation (48%). Moreover, fungi such as Phialophora sp., Cladosporium tenuissimun, and Aspergillus were isolated from surfaces of stone monuments [52] [53] [54].

Microbial metabolites enable some substances from rocks or minerals such as Si, Al, Fe, Mg, Mn, Ca, K, Na, Ti, to leach out from their salts, especially because of the impact of microorganisms on the dissolving rate of minerals [55].

Two bacterial isolates were positively identified as being capable of dissolving calcium carbonate, they belong to the genus Bacillus: B. cereus OK447647 and B. subtilis OK447648 (Figure 10). These results come in agreement with Abd-Elkareem and Mohamed (2017), who reported that calcium carbonate-dissolving Bacillus cereus, B. subtilis and B. circulans were recovered from deteriorated limestone of Sultan Hassan Mosque, Cairo-Egypt. On the other hand, Sonntag (2015) [56] isolated Brevibacterium sp. from Krast caves, this bacterium is reported their calcite dissolution ability. Moreover, three calcite dissolving bacteria (Bacillus megaterium, B. aryabhattai and Brevibacterium) were isolated from calcareous soil samples from Tamil Nadu, India [57]. Fungal isolates reported in this work were incapable of dissolving calcium carbonate. Morales et al. (2016) [58] reported that fungal isolates which were capable of solubilizing calcium by means of organic acid release, represented only 26% of fungi isolated from the surface of Mayan buildings at Yucatan, Mexico. Palmer and Hirsch (1991) [59] as well, stated that not all fungal species or strains are able to dissolve calcium carbonate.

Based on the results shown in Table 4, sodium azide of concentrations up to 50 ppm were ineffective against all tested bacteria and fungi. At 100 ppm, all tested

microorganisms were inhibited, the mean diameter of inhibition zone fluctuated between 8 and 15 mm. Therefore, it can be concluded that 100 ppm was the MIC of sodium azide to inhibit all the tested bacteria and fungi. Similar findings were reported by Abdelhafez et al. (2012), who concluded that 100 ppm of sodium azide was the best treatment to stop the growth of all microbial isolates recovered from surfaces of archeological marble located in Cairo, Egypt.