ASAE Conference Proceeding
This is not a peer-reviewed article.
Inactivation of Pathogenic and Indicator Organisms in Cattle Manure by Anaerobic Digestion: Assessment by The Methods of Cultivation and qPCR
M. Effenberger, M. Lebuhn, P. A. Wilderer and A. Gronauer
Pp. 083-090 in the Animal, Agricultural and Food Processing Wastes, Proceedings of the Ninth International Symposium, 11-14 October 2003 (Raleigh, North Carolina, USA), ed. Robert Burns. ,11 October 2003 . ASAE Pub #701P1203
Application of animal manure is restricted in water protection areas in Germany to avoid microbial contamination of drinking water supplies. The potential of a newly designed AD process scheme for the control of pathogens in cattle manure is being investigated in a joint project of Bavarian research institutions and local water suppliers. Experiments are being conducted in a pilot plant (total digester volume: 250 m3) located on a dairy cattle farm sized for the treatment of semi-liquid manure (7 to 10 % m/m dry matter) from 100 livestock units (approximately 2,000 m3/a). The plant consists of a sequence of three anaerobic digesters operating at different temperature levels (mesophilic-thermophilic-mesophilic – "MesTherMes") to optimize the destruction of a wide range of pathogenic organisms.
Using solely semi-liquid cattle manure as it was produced on the farm, the AD process in the pilot biogas plant could be established within about 2.5 months. Technical and operational problems particularly with the heating system prevented the plant from reaching steady-state conditions during the time period reported here. In preliminary experiments the quantitative real-time polymerase chain reaction (qPCR)-method in parallel with conventional cultivation techniques was successfully applied to determine the fate of indicator microorganisms in the pilot biogas plant. The required time for a complete qPCR-analysis was only 6-8 h as opposed to 24-72 h for cultivation. Higher values from the qPCR-analyses compared to cultivation assays for certain bacteria / bacterial groups indicated differences in specificity, and detection of microorganisms in a dormant, non-cultivable state by the qPCR-method.KEYWORDS. Cattle Manure; Pathogen Reduction; Pathogen Detection; Anaerobic Digestion
Potential causative agents of human waterborne infections that may be present in manure include bacteria, protozoa, and viruses (Bicudo et al., 2000). At the watershed level animal agriculture is considered a non-point or diffuse source for pathogens and undesired nutrients in the environment (Lenhart, 2001) . This can include runoff and drainage from grazing and farm land, drain pipes of farmyards, and leaking dunghills . Livestock farms have been found to be the source of microbial contamination of drinking water in a limited number of cases (Stehman, 2000). The (oo)cysts of the protozoan parasites Giardia sp. and Cryptosporidium parvum are highly resistant to environmental stresses and chlorine treatment, and can remain viable and infectious in water for up to several months (Robertson et al., 1994). Monitoring of raw or drinking water for Cryptosporidium and Giardia is not routinely done in Germany, and standards for these organisms have not been established (Botzenhart & Exner, 2001).
As a preventive measure, water protection areas are established around drinking water supplies in Germany. The aim of the so-called inner protection zone ("Zone II") is to avoid contamination of the drinking water especially by pathogenic microorganisms (DVGW, 1995). Both application and storage of animal manure are prohibited in this zone. Water supply companies have to refund farmers for disadvantages from land use restrictions. The water authorities are not willing to reconsider land use restrictions unless animal waste for land application has undergone a treatment that guarantees sufficient and reliable inactivation of relevant pathogens. A clear definition of “hygienically safe” animal manure, however, is still lacking. Treatment and examination of biological wastes other than sewage sludge for land application are regulated in the Ordinance on Biowastes (German: BioAbfV; Anon., 1998). Animal manure from clinically healthy livestock, however, is only amenable to these regulations if mixed with other biological wastes.
Beside generation of a versatile renewable energy source (biogas), anaerobic digestion (AD) offers other benefits such as recycling of nutrients, reduction of odor, and improvement of fertilizing effects (Wright, 2000). Also, it has been shown that AD causes the inactivation of various pathogens present in animal manure (e.g., Haas et al., 1995; Kearney et al., 1993; Olsen & Larsen, 1986). AD at mesophilic temperatures (around 35°C) does not provide sufficient destruction of pathogenic organisms within practicable treatment times (24 h or less). Thermophilic conditions (around 55°C) can reduce the numbers of several pathogenic bacteria, viruses and parasite eggs in liquid animal wastes by several orders of magnitude within hours, with temperature as the dominant inactivating factor. However, some spore-forming bacteria, certain viruses, and (oo)cysts of endoparasites are quite heat-resistant (e.g., Soares et al., 1994; Whitmore & Robertson, 1995). Cryptosporidia in sentinel chambers were not safely eliminated by anaerobic treatment of semi-liquid cattle manure in a single continuously–stirred tank reactor within 24 h at 55°C (Doll, 1999). Complete inactivation of these organisms was reported in the case of aerobic stabilization if they were first exposed to a phase of pre-heating the substrate (35 to 45°C) prior to treatment at thermophilic temperatures (>50°C) (Oechsner & Doll, 2000). It was concluded that passing through mesophilic temperature conditions improved the inactivation of Cryptosporidia by stimulating excystation of the heat-resistant oocysts.
In continuously-stirred tank reactors which are by far the dominant form used in agricultural biogas plants in Bavaria and Germany (Effenberger et al., 2002), the retention time is given by the time interval between withdrawal and feeding. From the hygienic point of view this type of digester requires long feeding intervals, which are not desirable with regard to process stability and continuous biogas production. To date long-term experiments to answer questions concerning the technical and operational requirements of agricultural biogas plants in order to maximize their sanitizing effect are lacking. One reason for this may be the fact that the culture-based techniques commonly utilized for quantification have several major drawbacks such as long analysis time, lack of specificity, or the fact that some pathogens such as certain viruses and endoparasites cannot be cultivated at all (Lebuhn et al., 2003).
The potential of a newly designed AD process scheme for the control of pathogens in cattle manure is being investigated in a joint project of Bavarian research institutions and local water suppliers. This paper gives a description of the pilot biogas plant designed to maximize the reduction of a wide range of pathogenic microorganisms. It presents preliminary data from the initial phase of operation of the plant and first results from the microbiological analyses of pathogenic and indicator organisms in cattle manure using the techniques of qPCR and conventional cultivation.
Materials and Methods
Pilot Biogas Plant
A sequence of three anaerobic digesters operating at different temperature levels (mesophilic-thermophilic-mesophilic – "MesTherMes") was chosen to achieve destruction not only of less problematic pathogens and commensals but also of the (oo)cysts of endoparasites and possibly bacterial spores. The goal is to maximize the sanitizing effect of the anaerobic treatment of cattle manure during quasi-continous operation (hourly feeding) and without a pasteurization step. The pilot plant is located on a dairy cattle farm and sized to treat approximately 2,000 m3 of semi-liquid manure per year.
Semi-liquid manure from a dairy cattle stable with slatted floors is collected in an underground canal, using automatic scrapers. The dairy cattle are fed with a mixture of grass silage and total mixed ration all the year-round. The semi-liquid manure (dry matter content: approximately 7 to 10 % m/m) is pumped into the collection tank in batches with an immersed chopping pump. From the collection tank the substrate is delivered to the first and subsequently to the second and third anaerobic digesters with progressive cavity pumps (feed rate: 4 m3/s). The MesTherMes-process consists of a stirred tank digester followed by a horizontal tubular digester and another, larger stirred tank digester. The first digester (total volume: 50 m3) was designed to activate (oo)cysts of protozoan parasites and possibly bacterial spores at mesophilic temperature conditions so that they can be inactivated by the subsequent thermophilic AD step. The tubular digester (12 m in length, 2.4 m in diameter, total volume ca. 46 m3) is equipped with an axial mixing device, and baffles are installed to reduce longitudinal mixing and maximize actual pathogen retention time during quasi-continuous operation. The third, mesophilic digester (150 m3) provides biological stabilization of the substrate. Flow of substrate is only allowed in one direction, and separate pumps are provided for successive treatment steps. Feeding, mixing, and heating of the digesters are operated automatically by a programmable logic controller. The digested manure overflows into a storage tank with a gas-tight cover in which the microbial status of the treated manure can be monitored over longer time periods while excluding microbial contamination from wildlife.
The biogas from digesters 1 and 2 flows into digester 3 which is covered by an elastic tarpaulin and thereby serves for biogas storage. Additional storage capacity for the biogas is provided by the covered storage tank. The produced biogas is conducted to the combined heat and power unit that is equipped with a pilot injection engine (Perkins; maximum electrical power output: 30 kW). To protect the engine from high sulfur levels in the gas, atmospheric air is introduced into the first digester by means of an aquarium air pump. The atmospheric oxygen is used by sulfide-reducing bacteria to produce sulfur (so-called biological desulfurization).
Figure 1. Schematic overview of the investigated manure treatment process
During the initial phase of operation of the pilot biogas plant from November 2002 to May 2003 continuous monitoring of process parameters included: substrate quantities using electromagnetic flow meters; temperatures at different points in the digesters using thermistors; quantity of combined biogas flows from the digesters and the storage tank using a drum-type gas meter; composition of combined biogas flows (nondispersive infrared sensors (NDIR) for methane and carbon dioxide; electrochemical sensors for hydrogen sulfide and oxygen).
Samples of raw and digested manure for chemical tests were taken from 5 points along the treatment process on a weekly basis. Liquid samples were analyzed for dry matter (DM), volatile solids (VS), chemical oxygen demand (COD), total alkalinity, volatile fatty acids (VFA), pH, and ammonia-nitrogen based on German standard methods (Anon., 1981).
Start-up and Initial Operation of Biogas Plant
The first and second digesters of the pilot plant were started up at the end of August 2002 using solely raw semi-liquid cattle manure that was diluted with water at a ratio of about 1:1. The third digester was subsequently filled with the effluent from the second digester. As long as there was no biogas production, the heating energy for the digesters was supplied by running the pilot injection engine on Diesel fuel. In the start-up phase the reactors were fed manually after total VFAs in the digesters had declined to levels of around 1000 mg acetic acid equivalent (aae)/L. Individual VFAs were consecutively monitored to adjust daily quantity of feed. Automatic quasi-continuous feeding at an interval of one hour was started in the middle of November 2002 with a daily quantity of 1.5 m3 (ca. 60 L/h) of semi-liquid cattle manure. The feeding was slowly raised to a value of approximately 6 m3/d (250 L/h) at the beginning of March 2003. It was subsequently adjusted to a value of 5.5 m3/d (230 L/h) according to the average daily production of semi-liquid manure in the dairy cattle stable.
The sanitary performance of the AD process was evaluated by monitoring the microbial parameters listed in Table 1, using classical (cultivation) and real-time quantitative polymerase chain reaction (qPCR, optionally with an upstream reverse-transcription, RT, step targeting RNA) techniques (Holland et al., 1991) in parallel. Available classical techniques were applied, and primers and probes for (RT)qPCR have been developed and tested for specificity. A major issue of the project was to adjust and optimize these techniques for working with dairy cattle manure. An optimized DNA extraction protocol, which involves standard spiking of cattle manure, was developed after comparing different DNA extraction protocols and testing the effect of sample washing prior to DNA extraction (Lebuhn et al., 2003).
Table 1: Evaluated microbial parameters during the initial phase of operation of the pilot biogas plant
Results and Discussion
Operational Data of the Biogas Plant
Roughly two months after the digesters had been initially filled the combined heat and power unit was run on biogas for the first time. Total VFA concentrations in samples taken from the digesters 1 and 2 shortly after filling and heating-up showed typical high levels of up to 5500 mg aae/L (Figure 2). After four weeks without feeding VFA concentrations in the digesters 1 and 2 had dropped to about 1300 and 1000 mg aae/L, respectively. A second pronounced peak of VFA levels after the start of feeding in November 2002 occurred only in digester 1. Acetic acid was the dominant component in most of the digester samples except in digesters 1 and 2 during the first weeks of start-up when propionate predominated (data not shown). The increased VFA levels in the samples of digesters 1 and 2 from mid-December can be attributed to a short-term decrease of temperature in the first and second digester by about 1 and 2 degrees, respectively. Total VFA levels in digesters 1 and 2 decreased to around 1000 mg aae/L within the following two months (Figure 2).
Data from chemical analyses of substrate samples that were taken once a week are reported here for the time from the end of January to the end of March 2003. Dry matter contents of raw and treated manure were 6.6 to 9.8 % m/m, 6.6 to 8.0 % m/m, 5.8 to 7.0 % m/m, and 4.4 to 5.3 % m/m in samples from the collection tank and digesters 1 to 3, respectively. Corresponding percentages of volatile solids (m/m) were 73 to 77 %, 72 to 74 %, 71 to 72 %, and 66 to 69 %, respectively. Concentrations of ammonia-nitrogen in samples from the three digesters were between 1.7 and 2.2 g/L. The pH-values in these samples measured immediately after sampling ranged between 7.2 and 8.0. Total alkalinity ranged from 250 to 320 mmol/L in samples of raw manure and from 300 to 350 mmol/L in samples from the digesters, respectively.
From mid-November 2002 until the beginning of March 2003 the methane content of the combined biogas flows from all tanks rose from about 30 Vol.-% to a level of about 53 to 54 Vol.-%. Latest measurements in May 2003 exhibited methane concentrations between 55 and 59 Vol.-%. Given the remaining concentrations of about 1 Vol.-% of oxygen from the biological desulfurization process this corresponds to a methane content of about 60 to 64 Vol.-%. Due to the very cold weather in February 2003 the biological desulfurization process basically came to a standstill, and measured H2S concentrations peaked at a value of 3839 ppm. With rising ambient temperatures during May 2003, H2S concentrations in the biogas found their level mostly below 60 ppm with remaining oxygen concentrations of around 1 Vol.-%.
Due to technical shortcomings particularly with respect to the heating system, the pilot biogas plant did not reach steady-state conditions during the reported time period. Temperatures in different sections of the thermophilic digester ranged between 45 and 56°C. Average daily biogas production in March 2003 at a daily feeding amount of around 6 m3 dairy cattle manure was 124 normal m3 corresponding to a specific biogas production of about 21 m3 per m3 of semi-liquid cattle manure. Data from the chemical analyses indicated no instabilities of the anaerobic digestion process. The average specific methane yield during March 2003 was 0.23 m3 methane per kg volatile solids at an average loading rate of 1.73 kg volatile solids/m3 related to the total volume of all three digesters (about 250 m3).
Figure 2: Concentrations of total volatile fatty acids (C2 to C5) in mg acetic acid equivalents/L in samples from digesters during start-up phase of the pilot biogas plant, as determined by gas chromatography
Preliminary experimental data for fecal coliform bacteria / Escherichia coli , and (intestinal) enterococci / Enterococcus faecium from the initial phase of operation of the biogas plant are presented for the treatment chain when the thermophilic reactor was operated at 48°C and 51°C, respectively. Table 2 shows a compilation of data for these organisms obtained by classical cultivation and qPCR. Using our optimized qPCR protocol (Lebuhn et al., 2003), we were able to prove the presence of 1 DNA target molecule in the reaction cavity and in 10 mL (waste) water samples (Burtscher & Lebuhn, 2002). However, since only 40 µL of cattle manure or digested material could be applied to the MiniPrep extraction kits, the theoretical detection limit was 250 organisms per 1 mL of cattle manure.
Table 2 shows equivalent results for fecal coliform bacteria and Escherichia coli in fresh manure, suggesting that the qPCR system may substitute the cultivation-based analysis. However, with increasing treatment time, qPCR data exceeded those of cultivation. This may be explained by differences in specificity between the compared systems, the fact that many target bacteria in the cattle manure may have been in a dormant, non-cultivable (ABNC) state, or that they were dead. Enterococcus faecium was dominant as compared to E. faecalis (not shown). For the enterococci, 10 - 1000 fold lower values were generally obtained by the cultivation approaches than by the corresponding qPCR analyses. The ABNC state is typical, e.g., for enterococci and campylobacters at suboptimal conditions (Lleò et al., 2001; Thomas et al., 2002). Presence and quantity of such cells may be proved by the use of reverse transcription of messenger RNA (mRNA) before qPCR (RTqPCR).
For the fecal coliform bacteria / Escherichia coli , and (intestinal) enterococci / Enterococcus faecium , a reduction of approximately 3 log units was obtained by the treatment chain under the described suptoptimal operational conditions (Table 2). Thermophilic treatment at 51°C was slightly more efficient than at 48°C. Preliminary experiments demonstrated the suitability of the (RT)qPCR methodology also for C. parvum DNA and enteroviral RNA. Further experiments are required to demonstrate viability of the targeted quantified organisms, and to optimize extraction of less stable RNA from cattle manure. The time necessary to perform the complete analyses was 6 - 8 h for the qPCR approaches (including DNA extraction) and 24 - 72 h for the cultivation based systems.
Table 2: Cultivation-based and qPCR analyses of microorganisms / microorganism groups in fresh and digested cattle manure.
Conclusions and Outlook
Despite technical and operational shortcomings the anaerobic digestion process in the pilot biogas plant was started up without problems and has so far exhibited no severe instabilities. Real-Time Quantitative PCR (qPCR) appears to be highly suited to determine the fate of specific organisms or pathogens during anaerobic digestion and is an attractive alternative to the culture based systems for quantification of (indicator) organisms. As soon as steady-state conditions in the biogas plant will have reached, the optimized qPCR-protocol will be used to examine and optimize the performance of the pilot plant for the inactivation of pathogens in cattle manure, including particularly resistant organisms. Additional process data relating to substrate degradation, biogas production, energy and economic efficiencies will be collected over the course of at least one year. The data will contribute to an improved evaluation of the potential of AD in agriculture for minimizing environmental hygienic risks from livestock farming. It will serve as a basis for considerations whether this treatment could be an environmentally sustainable and economical alternative to the current practice of refunding farmers for restrictions of manure application to land.
This work is funded by Bayerisches Staatsministerium für Landwirtschaft und Forsten and Bayerisches Staatsministerium für Landesentwicklung und Umweltfragen.
Bicudo, J. R., S. M. Goyal, and J. Zhu 2000. Animal Production, Manure Management and Pathogens: A Review. In Animal, Agricultural and Food Processing Wastes, Proceedings of the Eighth International Symposium, 507-521 . J.A. Moore ed., St. Joseph, Mich.: ASAE.
Lenhart, B. 2001. Bedeutung und Verminderung diffuser Belastungen im Einzugsbereich "Obere Isar". In Hygienische Aspekte von Oberflächengewässern aus wasserwirtschaftlicher Sicht , 167-175. Munich, Germany: Bayerisches Landesamt für Wasserwirtschaft.
Stehman, S. M. 2000. Ag-Related Waterborne Pathogens. In Managing Nutrients and Pathogens from Animal Agriculture , 93-107. Ithaca, New York: Natural Resource, Agriculture, and Engineering Service (NRAES).
Robertson, L.F., H. V. Smith, and J. E. Ongerth 1994. Cryptosporidium and Cryptosporidiosis. Part III: Development of water treatment technologies to remove and inactivate oocysts. Microbiol. Europe 1/2:18-26.
Botzenhart, K. and C. Exner 2001. Health related microbiology: Dealing with waterborne Cryptosporidium and Giardia and other waterborne pathogens. German National Report for the IWA World Water Congress Berlin 2001: http://www.dvgw.de/pdf/iwa_microbio.pdf. Accessed on 10 Dec. 2002.
DVGW 1995. Richtlinien für Trinkwasserschutzgebiete; I. Teil: Schutzgebiete für Grundwasser. Arbeitsblatt W 101. Deutscher Verein des Gas- und Wasserfaches (DVGW) (German Association on Gas and Water).
Anonymous 1998. Ordinance on the Utilisation of Bio-Wastes on Land used for Agricultural, Silvicultural and Horticultural Purposes (BioAbfV) as of 21.09.1998 (BGBl. I 1998 S. 2955).
Wright, P. 2000. Manure Treatment and Handling Options. In Managing Nutrients and Pathogens from Animal Agriculture , 156-173. Ithaca, New York: Natural Resource, Agriculture, and Engineering Service (NRAES).
Haas, B., R. Ahl, R. Böhm, and D. Strauch 1995. Inactivation of viruses in liquid manure. Rev. sci. tech. Off. int. Epiz. 14 (2): 435-445.
Kearney, T. E., M. J. Larkin, and P. N. Levett 1993. The effect of slurry storage and anaerobic digestion on survival of pathogenic bacteria. J. Appl. Bacteriol. 74: 86-93.
Olsen, J. E. and H. E. Larsen 1986. Bacterial decimation times in anaerobic digestions of animal slurries. Biological Wastes 21: 153-168.
Soares, A. C., T. M. Straub, I. L. Pepper, and C. P. Gerba 1994. Effect of anaerobic digestion on the occurrence of enteroviruses and giardia cysts in sewage sludge. J. Environ. Sci. Health A29(9): 1887-1897.
Whitmore, T. N. and L. J. Robertson 1995. The effect of sewage sludge treatment processes on oocysts of Cryptosporidium parvum. J. Appl. Bacteriol. 78: 34-38.
Doll, L. 1999. Verfahrenstechnische Untersuchungen zur Entseuchung von Flüssigmist mittels thermophiler Faulung und aerob-thermophiler Stabilisierung. PhD diss. VDI-MEG publication 345. Stuttgart: Univ. Hohenheim.
Oechsner, H. and L. Doll 2000. Inactivation of Pathogens by Using the Aerobic-Thermophilic Stabilization Process. In Animal, Agricultural and Food Processing Wastes, Proceedings of the Eighth International Symposium, 522-528 . J.A. Moore ed., St. Joseph, Mich.: ASAE.
Effenberger, M., M. Schlattmann, F. Kaiser and A. Gronauer 2002. Stand der Biogastechnologie in Bayern. In Biogasanlagen – Anforderungen zur Luftreinhaltung . Augsburg: Bavarian Environmental Protection Agency.
Lebuhn, M., M. Effenberger, A. Gronauer and P. Wilderer 2003. Using quantitative real-time PCR to determine the hygienic status of cattle manure. Water Sci. Tech. In press.
Anonymous 1981. Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung. Wasserchemische Gesellschaft, Fachgruppe in der GDCh / in Gemeinschaft mit dem Normenausschuss Wasserwesen (NAW) im DIN e.V. (eds.). Weinheim: Wiley-VCH.
Holland P. M., R. D. Abramson, R. Watson, and D. Gelfand 1991. Detection of specific polymerase chain reaction product by utilizing the 5' ? 3' exonuclease activity of Thermus aquaticus DNA polymerase. Proc. Natl. Acad. Sci. USA 88: 7276-7280.
Burtscher, C. and M. Lebuhn 2002. Quantitative detection of Salmonella spp. in environmental samples using real time PCR. BIOspektrum , Sonderausgabe 2002: 62.
Lleò, M.M., B. Bonato, M. C. Tafi, C. Signoretta, M. Boaretti, and P. Canepari 2001. Resuscitation rate in different enterococcal species in the viable but non-culturable state. J. Appl. Microbiol. 91: 1095-1102.
Thomas, C., D. Hill, and M. Mabey 2002. Culturability, injury and morphological dynamics of thermophilic Campylobacter spp. within a laboratory-based aquatic model system. J. Appl. Microbiol. 92: 433-442.