2.0 Definition
Bacterial production is the rate of synthesis of biomass by heterotrophic bacterioplankton, as estimated by the incorporation of ³H-methyl thymidine into the cold trichloroacetic acid-insoluble and cold ethanol-insoluble cell fractions following a short term incubation, using a suitable conversion factor, F:
        Bacterial production (cells kg ^-1 h ^-1 ) = F*[³ H-thymidine] pmole kg^-1h^-1
        F = production of bacterial cells/mole ³H-thymidine

3.0 Principle of analysis
The rate of bacterial production is estimated by tracing the specific incorporation of ³H-thymidine into the TCA-insoluble macromolecular fraction. The incubation is terminated by adding formalin, followed by an extraction of the unincorporated ³H-thymidine from the bacterial cells in cold TCA and ethanol.

4.0 Apparatus
4.1 Filtration Apparatus. The tritiated incubation solution can be filtered using any reliable, leak-free, acid-resistant, multi-place filtration unit.

4.2 Liquid Scintillation Analyzer. Samples in liquid scintillation cocktail are counted on a liquid scintillation analyzer, using the following energy window settings:
        Channel A: 0-19 KeV
        Channel B: 2-19 KeV
Samples should be counted long enough to reduce the counting error to <5-10%.

4.3 Quench Corrections. An external gamma source is used to assess quenching of individual  filter samples for conversion of counts per minute (CPM) to disintegrations per minute (DPM). Quenching of the total radioactivity vials is determined by an internal standard (usually tritiated water added diluted concentrations of toluene or chloroform as a quencher).

5.0 Reagents
5.1 Stock of methyl-³H-thymidine (approximately 80 mCi/mmol) is stored in 96% ethanol in the refrigerator. Stock solution should not be frozen.

5.2 Working solution. An aliquot of the stock solution is transferred to a glass vial where  the ethanol is evaporated. The evaporation is promoted by a vacuum pump drawing air through a Silicagel-cartridge and a 0.2 mm Nuclepore filter. The tritiated thymidine is redissolved in 0.2 mm filtered Milli-Q water (1 mCi/5 ml Milli-Q) and stored in the refrigerator not longer than 1-2 days before being used.

5.3 Acid Cleaning Solution (1N HCl Baker Analyzed) is prepared using Milli-Q water.

5.4 Incubation bottles. Polycarbonate centrifuge tubes (29 ml) are used for the bacterial productivity incubations. Before every cruise, the tubes are soaked in KOH, rinsed in Milli-Q water and finally soaked in the acid solution overnight. The acid is then discarded and the tubes are rinsed and soaked in Milli-Q water overnight. The polycarbonate tubes are emptied (remaining Milli-Q water is shaken out) and air-dried.

5.5 Concentrated (37%) formaldehyde

5.6 Trichloroacetic Acid (TCA) is made up in a 5% solution (weight/volume) in Milli-Q water. A premixed 100% TCA solution can also be purchased and diluted to a 5% working solution. The working solution is kept at 4 ° C in the refrigerator. Great care should be taken when working with dry or 100% TCA.

5.7 Ethanol (96%) is kept at 4 °C in the refrigerator.

5.8 Ethyl acetate (Purified, Baker Analyzed)

5.9 Liquid scintillation cocktail. Aquasol (New England Nuclear) or equivalent formulations  provide high efficiency counting of low-energy tritium beta particles. Non-toxic, biodegradable scintillation cocktails are now required by some institutions. Ultima Gold (Packard) provides results comparable to Aquasol if cellulose nitrate filters are completely dissolved in ethyl acetate prior to addition of cocktail. Other filter-cocktail combinations should be tested before substitution for those recommended in this manual.

5.10 Preparation of Reagents and Incubation vessels. Polyethylene gloves are worn during handling of all materials that are being used for the incubation.

6.0 Sampling and incubation
6.1 Sample dispensing. Polyethylene gloves are worn during sampling and all subsequent  manipulations. The polycarbonate centrifuge tubes are filled directly from the Go-Flos and rinsed 3 times before filling. Three centrifuge tubes are filled from each depth and stored in the dark during sampling. Several killed blanks from different depths should also be prepared. Samples for the estimation of bacterial abundance (see Chapter 18) should be taken at the same time.

6.2 Isotope inoculation. Under low light conditions, 100 ml of the tritiated thymidine working solution is added to each tube to a final concentration of about 10 nM. Ideally, samples should be inoculated and incubated at in situ temperatures. This can be accomplished using temperature-controlled, refrigerated incubators and/or flowing seawater-cooled incubators. The incubation should last sufficiently long to obtain measurable uptake but not so long as to cause uptake to depart from linearity. This may need to be determined for new habitats or depths. 1-2 hours is usually sufficient for most samples less than 200 m.

6.3 Time zero samples are made from triplicate aliquots of 20 ml of seawater from several depths.The aliquots are terminated by adding 200 ml concentrated (37%) formalin, followed by the addition of 50 ml tritiated thymidine working solution. The solutions are immediately filtered and extracted as described in section 7.1 of this chapter.

6.4 End of Incubation. The incubation is ended by subsampling aliquots of 20 ml by syringe from each tube into a separate reagent tube containing 200 ml concentrated (37%) formalin. The aliquots are immediately filtered and extracted as described in section 7.1 of this chapter.

7.0 Procedures
7.1 Filtration and extraction.Under low light conditions, the sample aliquots are filtered onto 25 mm diameter Sartorius (or MFS) cellulose nitrate, 0.22 mm pore size filters, maintaining a vacuum pressure of 70 mm Hg or lower. Mixed esters should not be used as they bind DNA and result in insufficient counting. If care is taken in emptying the reagent tubes, further rinsing is not necessary. After the filter funnel is removed, and with the vacuum pressure maintained, the filters are rinsed with 3 rinses of ice-cold 5% TCA solution from a wash bottle. The TCA rinses are followed with 3 rinses of ice-cold ethanol from another wash bottle. The wash bottles should be kept cold in an ice bucket filled with crushed ice and water during the filtration operation. Care should be taken to rinse the outer edges of the filters.

7.2 Filter processing and counting. The filters are placed in glass scintillation vials and allowed to dry completely overnight. If 7 ml scintillation vials are used, the filters need to be folded carefully 3 or 4 times so they are small enough to permit full immersion in the ethyl acetate. 0.5-1 ml ethyl acetate is added to dissolve the filters. Failure to dry or fully cover the filters in the ethyl acetate solution may result in incomplete dissolution and poor counting efficiency. Vortex mixing can be employed to aid in dissolving the filters, Finally, when the filter solution is clear, liquid scintillation cocktail is added, the cocktail solution is mixed and the samples are counted on a liquid scintillation counter.

7.3 Total Radioactivity Sample. Aliquots of 50 ml from three random incubation tubes are added to a set of three scintillation vials with 10 ml of scintillation cocktail to determine the total amount of label added to the samples.

8.0 Calculation and expression of results
Rate calculations. Universal factors for conversion of ³H-thymidine incorporation into cell production do not exist (Kirchman et al. 1982; Ducklow and Carlson 1992) but there is fair consensus that a the conversion factor (F) varies in the coastal and open ocean within 2±2 x 10¹⁸ cells mole^-1 . The rate of incorporation is reported as pmole 3 H-thymidine taken up per time unit after zero-time blank values are subtracted.
        [methyl-³H-thymidine] pmole kg ^-1 h ^-1 = (DPM/2200)*(1000/V)*(1/SA)*(60/T)

Where:
        DPM = disintegrations per minute of sample minus blank value
        V = extraction volume (20 ml)
        SA = specific activity (of added ³H-thymidine)
        T = incubation time (min)
A check on the final concentration of the tritiated incubation solution is estimated by converting the amount of the measured total activity into the final concentration of tritiated thymidine.
        [methyl-3 H-thymidine] nM = (DPM/2200)*(1000/ml) * (1/SA)
Where:
        ml = aliquot taken from incubation solution (50 ml)
        SA = specific activity

9.0 Quality Control
9.1 Standards and precision. There is no absolute standard for bacterial production measurements and the accuracy is unknown. The coefficient of variation of assays performed carefully following this protocol should be 15-20% for triplicate incubations. The limit of detection will vary depending on length of incubation and the amount of sample filtered. With care, incorporation rates of 0.05-0.1 pmol l ^-1 h ^-1 should easily be detected above background.

9.2 Non-specific incorporation of thymidine. Much of the uncertainty with thymidine results appears due to non-specific labelling. Tritiated thymidine does not seclusive enter the bacterial DNA and several studies have demonstrated the labelling of macromolecular compounds other than DNA (Hollobaugh 1988). Non-specific labelling makes it very important to use an extraction procedure specific for tritiated DNA (Wicks and Robarts 1987, Hollibaugh 1988, Robarts and Wicks 1989). New techniques using enzymatic digestion (Torreton and Bouvy 1991) also look promising.

10.0 Interpretation of results
A conversion factor is needed to derive bacterial production (cells or mass of C or N produced per unit time) from the incorporation rates. Conversion factors should ideally be determined experimentally for each new environment or season sampled. To determine a conversion factor, an independent measurement of bacterial production or growth rate must be made, or the relationship between thymidine incorporation and production must be determined. A variety of approaches exist for this purpose (Bjørnsen and Kuparinen, 1991; Ducklow et al., 1992; Kirchman and Ducklow, 1993). For open ocean sites the conversion factor is generally 2±2x10¹⁸ cells produced per mole incorporated.

11.0 References
Bjornsen, P.K., and J. Kuparinen (1991). Determination of bacterioplankton biomass, net production and growth efficiency in the Southern Ocean. Mar. Ecol. Prog. Ser. 71:185- 194.
Carman, K.R., F.C. Dobbs and J.B. Guckert. (1988). Consequences of thymidine catabolism for estimates of bacterial production: An example from a coastal marine sediment. Limnol. Oceanogr. 33:1595-1606.
Ducklow, H.W. and C.A. Carlson. 1992. Oceanic Bacterial Production. Advances in Microbial Ecology. K.C. Marshall ed. p. 113-181. Plenum Press.
Ducklow, H.W., D.L. Kirchman and H.L. Quinby. (1992). Bacterioplankton cell growth and macromolecular synthesis in seawater cultures during the North Atlantic spring phytoplankton bloom, May 1989. Microb. Ecol. 24:125-144.
Fuhrman, J.A. and F. Azam. (1980). Bacterioplankton secondary production estimates for coastal waters of British Columbia, Antarctica and California. Appl. Environ. Microbiol. 39:1085-1095.
Fuhrman, J.A. and F. Azam. (1982). Thymidine incorporation as a measure of heterotrophic bacterial production in marine surface waters: evaluation and field results. Mar. Biol. 73:79-89.
Hollibaugh, J.T. (1988). Limitations of the [ 3 H]thymidine method for estimating bacterial productivity due to thymidine metabolism. Mar. Ecol. Prog. Ser. 43:19-30.
Kirchman, D., H. Ducklow and R. Mitchell. (1982). Estimates of bacterial growth from changes in uptake rates and biomass. Appl. Environ. Microb. 44:1296-1307.
Robarts, R.D. and R.J. Wicks. (1989). Methyl-3 H thymidine macromolecular incorporation and lipid labeling: their significance to DNA labeling during measurements of aquatic bacterial growth. Limnol. Oceanogr. 34:213-222.
Torreton, J.P. and M. Bouvy. (1991). Estimating bacterial DNA synthesis from [ 3 H]thymidine incorporation: Discrepancies among macromolecular extraction procedures. Limnol. Oceanogr.36:299-306.
Wicks, R.J. and R.D. Robarts. (1987). The extraction and purification of DNA labelled with methyl-3 H thymidine in aquatic bacterial production studies. J. Plank.Res. 9:1159-1166.