Growth Rate and Histamine Production of Citrobacter freundii CK01 in Various Incubation Temperatures

Improper handling and temperature fluctuations during postharvest of fish commonly allow the growth of histamine-forming bacteria (HFB) that may cause histamine poisoning. Citrobacter freundii CK01 is one of the HFB isolated from Skipjack landed on Sadeng, Yogyakarta. This study aimed to determine the temperature effect on the growth rate and histamine production by C. freundii CK01. Bacterial growth and histamine production was tested in tuna fish infusion broth (TFIB) at 5, 15, 30 and 40°C. The bacterial growth was tested using Total Plate Count method, while histamine was determined using Thin Layer Chromatography and densitometry method. The primary model for bacterial growth was plotted with incubation time using DMFit referred to Baranyi & Roberts model. The secondary model was converted from growth rate and modeled using Ratkowsky Square Root Model. The equation for growth rate of C. freundii CK01 was μmax=[0.0105 (T + 13.886)] (Root Mean Square Error <10%). Histamine production reached the highest concentration at 15°C in 96 hours up to 117,13 ppm. This study shows that temperature affected the growth rate and histamine production of C. freundii, indicating the importance of maintaining the low temperature stability during handling of skipjack.


Introduction
Indonesia has a great potential in fisheries, one of the main commodities is skipjack tuna. The increasing number of Skipjack's traffic in Indonesia from 2014-2017 grew by an average of 52.81% per year and reached 29.04 thousand tons indicating an increase in the consumption and trade distribution activities of skipjack in Indonesia [1]. Skipjack is the second largest commodity caught in Gunungkidul Yogyakarta and all year's catching season [2]. Skipjack contains high nutritional component: 71.75% water; 25.29% protein; 1.49% ash; 0.59% fat and 0.8% carbohydrate [3].
To maximize the potential value of Skipjack, products have to meet the quality standards. Histamine content in Skipjack is one of the quality parameter that needs to be considered not to exceed the limit in order to avoid poisoning. Histamine is a chemical that is toxic if found in large quantities in the body. The histamine poisoning symptoms are dizziness, stomach nausea, vomit, rapid heart rate, constant thirst, and itchiness [4]. At the end of 2017 until the beginning of 2018, the United States Food and Drug Administration (US-FDA) sent a warning letter to 6 Indonesian fisheries companies related to HACCP in controlling food safety hazards from histamine formation [5]. In Indonesia, the extraordinary events of food poisoning (KLB) in 2000-2015 amounted to 61,119 cases and histamine poisoning contributed to 6.7% cases, whereas most cases were caused by pathogenic bacteria (74.9%) [6]. Histamine from sardine consumption was also reported by the Lhokseumawe Health Office as a source of poisoning to 72 high school students [7].
The formation of histamine in fishery products are commonly due to the bad distribution chain, less attention of sanitation-hygiene and improper cold chains practice, resulting in temperature fluctuations. Improper handling temperatures can cause histamine-forming bacteria (HFB) growth and histidine decarboxylase enzymes to become active [8]. HFB produces the L-histidine decarboxylase enzyme, which converts free-histidine in fish to histamine [9].
Citrobacter freundii was categorized as the HFB with the low ability of histamine production, since among the 152 pure bacterial strains cultured in Triptone Soya Broth added with L-hystidine, C. freundii only produced less than 500 ppm of histamine along with Hafnia [10]. C. freundii as one of the HFB in fish has not been widely studied to determine its potential as a histamine-causing agent. C. freundii CK01 originated from Skipjack in Sadeng Beach fish market was isolated in Niven's medium at 37°C [11]. The production of histamine by C. freundii CK01 in Tryptone Soya Broth + 0,1% L-histidine (TSBH) reached 1,600 ppm after 24 h of incubation at 37°C [11]. A similar study reported that C. freundii produced histamine up to 474 ppm in chub mackerel sample after 48 h incubation at 30°C [12]. This study was conducted to characterize the growth of C. freundii CK01 at various temperatures and its ability in producing histamine. It is expected to provide information to minimize the risk of scromboid poisoning in Skipjack related to storage temperature controlling.

Medium preparation
Tuna Fish Infusion Broth (TFIB) was prepared from fresh Katsuwonus sp. obtained from Sadeng Beach, Gunung Kidul, Yogyakarta. They were brought to the laboratory with styrofoam-box containing ice cubes, degutted, washed, then stored in the freezer overnight. The flesh were separated from the bone, washed, and homogenized with a blender in a water ration as twice as the weight of fish, then boiled at 100 °C for 1 h. The mixture was cooled, filtered with Whatman No.1 filter paper, added with 1% glucose, put in a test tube of 10 mL each, and sterilized [13]. Tryptic Soy Agar (TSA) (Oxoid) and Tryptic Soy Broth (TSB) (Oxoid) were prepared following the manufacture's protocol.

Inoculum preparation
Citrobacter freundii CK0; a bacterial collection from The Laboratory of Fisheries Product Quality and Safety, Department of Fisheries, Faculty of Agriculture UGM; in glycerol stock was streaked on TSA medium and incubated for 24 h at 37°C. The single colony was transfered into TSB and incubated for another 24 h at 37°C. The TSB culture was used to prepare working culture on TSA slant. Inoculum was prepared by transfering cells using inoculating needle from working culture in to TSB, followed by incubation for 24 h at 37°C.

Citrobacter freundii CK01 growth in various temperature
One loop of C. freundii CK01 inoculum prepared in TSB were cultured in TFIB and incubated at four temperature treatments: 5, 15, 30 and 40°C. Each treatment was done in two replicates. The temperature of 5 and 15°C were observed every 24 h for 5 days, while the temperature of 30 and 40°C every 3 h for 24 h. During the observation, cell growth and histamine production were determined. Cells growth were determined by calculating the cell number in the range of incubation time using the total plate count method. A 100 μL C. freundii CK01 culture in TFIB from each treatment was transfered into 900 μL Butterfield's phosphate buffer, homogenized, and made into serial dilutions. Each dilution was inoculated on TSA and incubated at 37 ± 2 °C for 24 ± 2 h before the calculation of cell number (log CFU/ml).
The histamine content produced in culture medium was determined semi-quantitatively using Thin Layer Chromatography (TLC) following previous method [14] with a modification in the spot analysis method. A 100 μL bacterial culture in TFIB was harvested and centrifuged at 13,000 rpm for 10 minutes. Samples (5 μL) were spotted on a 10 x 20 cm TLC plate 60 F254 (Merck KGaA Germany). The plate was inserted into a chamber containing mobile phase solvent prepared from a mixture of ammonia and methanol with a volume ratio of 3:1. After the migration process completed, the plate was air dried and sprayed with a solution of ninhydrin (300 mg ninhydrin in 100 mL n-butanol added with 3 mL of glacial acetic acid) to develop spots. The histamine solutions (100 to 1,000 ppm) prepared by diluting Histamine powder (TCI, Germany) in 1 M HCl were used as standard. The same procedure was applied for the L-histidine (TCI, Germany) standard solutions (1,000 to 10,000 ppm). The standard solutions were also spotted on to the TLC plate and following the same steps with the samples. The TLC plate was scanned and spots were analyzed to determine their area values using ImageJ program. The area of standards were plotted with the known concentration of standards resulting to the linear regression equation that were used to convert the area of the samples into the histamine or histidine concentration (ppm).

Data analysis
The data of cell number in each incubation time were used to obtain growth curve at various temperatures with an Excel program. The primary model of bacterial growth was obtained using the DMFit software (http://www.combase.cc) based on Baranyi   The growth of C. freundii CK01 in TFIB at 5, 15, 30 and 40°C are shown in Figure 1.  [16]. The number of C. freundii at 30°C increased significantly, from 7.1 to 9 log CFU/g after 48 h incubation, while at 37°C increased from 7 to 8.6 log CFU/g. They also showed that the number of C. freundii incubated for 72 h at low temperatures (-3 and 0°C) increased in a small amount, from 7 to 7.5 log CFU/g. Meanwhile, at 15°C, C. freundii showed fairly increase in the first 24 h, from 7.2 to 8.8 log CFU/g. The cell growth data of C. freundii CK01 were analyzed using DMFit to determine the bacterial growth rate (μmax). Overall, the μmax increased with the temperatures. The smallest value of μmax was observed at 5°C, followed by 15°C, 30°C, and the highest μmax was at 40°C (Table 1)  C. freundii optimally grows at 37°C [17]. Above the optimal temperature, denaturation of protein occurs resulting to the cellular disfunction, whereas under optimal temperature, bacteria lost their membrane function resulting to the disruption of bacterial metabolism [17]. The relationship between temperature and bacterial μmax was determined and modeled in the temperature functions using Ratkowsky Square Root Model (RSRM) [24]. The mathematical equation describing the relationship of temperature and bacterial growth rate square root of C. freundii CK01 was y = 0.0105x + 0.1458 with R²=0.9235 ( Figure 2). A slight decrease in growth rate square root was observed at 40°C. It indicates that 40ºC was slightly above the optimum temperature for M. morganii growth. A similar finding was found in other study which showed that the square root value of M. morganii growth rate increased with increasing temperatures up to 40ºC, but at a temperature of 45ºC the growth rate square root decreased [18]. The growth prediction model of C. freundii CK01 in the function of temperature was μmax = [0.0105 (T + 13.886)] 2 . Tmin is the minimum temperature when the growth rate is equal to 0. The Tmin value of C. freundii CK01 obtained by this model was lower than other HFB isolated from the same fish, Klebsiella sp. CK02, which produced Tmin of -8.89°C [11]. The lower the Tmin value, the lower the temperature needed to deactivate bacterial growth. The ability of C. freundii grown at 0°C was reported by the increasing of the cell number from 7 to 7.5 log CFU/g after 158 h incubation [16]. It indicates that C. freundii CK01 had a considerable growth resistance in low temperature conditions. This prediction model is useful to determine shelf life, optimize storage conditions, and for further study of histamine exposure assessments [18].

Histidine and histamine production by C. freundii CK01 in TFIB
The histidine standard was detected on the TLC plat as the pink color spot with the Rf value of 0.56, whereas histamine standard was dark red with the Rf value of 0.26 ( Figure 3). The histamine standards below 100 ppm and above 1,000 ppm was failed to be read as the ImageJ program was not able to distinguish the area values of the spots. The linear regression equation of y = 0.0008x + 0.0898 (R 2 = 0.9021) was used to estimate the histamine concentration, meanwhile the equation of y = 0.0001x -0.1645 (R 2 = 0.9363) was used to estimate the histidine concentration in bacterial culture medium, whereas y was the area value. Histidine concentration in the medium and histamine formation by C. freundii CK01 are shown in Figure 4. At 5 and 40 ºC, the histamine was not detected and histidine concentration did not decrease, indicating that histamine was not produced during the experiment or produced in a low level below the detectable range. C. freundii was reported to have no significant histamine production after incubation at 7ºC for 158 h [16]. C. freundii CK01 at 5 and 40°C showed a lower growth rate and the histamine production was not detected (below 100 ppm). The formation of high histamine concentration were detected at 15 and 30°C. The highest histamine formation was shown at a temperature of 15°C with an initial histamine of 94 ppm at 48 h to 117.13 ppm at 96 h. At 30°C the histamine formation reached 103.38 ppm within just 12 h. The optimal temperature of histamine production by C. freundii was reported at 37ºC, reaching 220 ppm after 72 h incubation in TSBH [16]. High histamine formation was also reported for C. freundii which produced 474 ± 9 mg/L (equivalent to ppm) histamine after 48 h incubation at 30ºC in TSBH [12]. The results in this experiment tend to show less histamine levels produced by C. freundii compared to the other two previous studies [16,12]. This was probably because of the different medium that were used. The production rate of histamine is affected by the free-histidine availability in the medium [21]. The higher the substrate concentration, the higher the product is produced, the histidine concentration as the substrate will affect the amount of histamine formed [19]. This explained the abundance of histamine found in Scromboidae fishes is due to their high content of free histidine in the muscle tissues [12]. The optimal temperature for histamine formation by C. freundii CK01 was different from the temperature of the highest growth rate. The highest histamine formation was produced at 15ºC in 96 hours of incubation while the highest growth rate of bacteria was achieved at 40ºC in 24 hours incubation time. This result indicates that handling of skipjack at 15ºC for a long period was potential for histamine accumulation. Moreover, the data shows that a high number of HFB in the sample was not directly related to a high concentration of histamine in the sample. Presumably the optimal temperature for bacterial metabolism that affected growth is different from the optimal temperature of the hitidine decarboxylation enzyme works. Similar trend was reported in Proteus morganii which showed a temperature difference for an optimal growth (37ºC) with the highest histamine production (24 and 30ºC) [20]. Another reported was in Photobacterium phosphoreum YS4-7 which showed the highest growth at 27ºC while the highest histamine concentration was produced at 20ºC [23].

Conclusion
This study shows the effect of temperature on the growth rate and histamine production ability of C. freundii CK01. Temperature fluctuations that occur during the process of skipjack tuna handling allow growth of C. freundii CK01 with varying growth rates that could be calculated by the equation μmax = [0.0105 (T + 13.886)] 2 . The most important finding in this study was C. freundii CK01 showed growth, although in a very low rate, at low temperature 5ºC. It indicates that the cell number might increase to a significant level when the low temperature handling proceeds for long period. Eventhough C. freundii CK01 did not produce a significant level of histamine at 5ºC, the high initial cell number in the fish due to its ability to grow at low temperature during handling might resulted to the high rate of histamine formation when there is a temperature abuse. The recommended fish storage temperature is determined through the Tmin value as the minimum temperature for bacterial growth, from this study it is recommended that safe storage temperature for Skipjack is around -13ºC to prevent histamine production. This study illustrates the importance of striving for low temperature stability and a short time in handling fresh skipjack.