does adenine incorporation into nucleic acids measure total microbial production?

Limnol. Oceanogr., 31(3), 1986, 627-636 0 1986, by the American Society of Limnology and Oceanography, Inc.

Does adenine incorporation into nucleic acids measure total microbial production?’

Jed A. Fuhrman Marine Sciences Research Center, State University of New York, Stony Brook 11794

Hugh W. Ducklow Horn Point Environmental Laboratories, University of Maryland, P.O. Box 775, Cambridge 21613

David L. Kirchman Department of Molecular Genetics and Cell Biology, The University of Chicago, 5630 S. Ingleside, Chicago, Illinois 60637

John Hudak and George B. McManus Marine Sciences Research Center, State University of New York, Stony Brook

Jonathan Kramer Horn Point Environmental Laboratories, University of Maryland

Abstract

Incorporation of radioactive adenine into DNA has recently been used as a measure of total microbial production in marine environments, sometimes with unexpected results. We have examined two of the assumptions on which the method is based, particularly regarding the distribution of adenine uptake and nucleic acid content in mixed natural microbial assemblages. Size fractionation, autoradiography, and metabolic inhibition data indicate that the requirement that bacteria, algae, and protozoa have a uniform uptake and metabolism of added dissolved adenine dots not appear to hold. In systems where the biomass and productivity were greatly dominated by large phytoplankton, bacteria took up almost all of the adenine added at nanomolar concentrations. A separate experiment with marine ciliates also showed adenine uptake by eucaryotes to be negligible compared to that by bacteria, The method requires a measurement of microbial adenosine triphosphate (ATP) specific activity. In our experiments this quantity would be dominated by ATP in the eucaryotes that took up negligible adenine. In this situation, mostly eucaryotic biomass and not production would influence the “total microbial production” measurement. Second, the C: DNA ratio of 50 used in reports to date is too high for natural bacteria, which have a ratio closer to 5. We conclude that the method is likely to have produced large overestimates of production and that variations (e.g. with depth) may not reflect growth rates, but rather a complicated com- bination of activities and ratios of eucaryotic : procaryotic biomasses.

Within the past year or two, some unexpected and potentially important productivity measurements have been reported for various marine environments, including the water column (both suspended and on sinking particles) and sediments. These reports share the same adenine incorporation meth- odology aimed at measuring the combined production of all bacteria, algae, and protozoa in the habitats studied. Specifically, Winn and Karl (19 84a) reported that total

’ This work was supported in part by NSF grants OCE 84-10074 and OCE 84-067 12 to J.F. and OCE 84-006524 to H.D., and by the NSF-sponsored MEC- CAS program.

microbial production at the 150-900-m depth of the waters off Hawaii is double that of the O-l 50-m depth range (euphotic zone). Burns et al. (1984) estimated total microbial production in coral reef rubble of Kaneohe Bay, Hawaii, and found rates higher than previous estimates of whole reef production. Karl and Knauer (1984) and Karl et al. (1984) reported an increase in total microbial production within rapidly sinking material at the 700-900-m depth and con- cluded that it is due to chemolithotrophic production. We believe that some of the key assumptions of the adenine method have not been tested adequately on natural samples under field conditions. Here we ex-

627

628 Fuhrman et al.

amine some of these assumptions with experiments on surface samples from estuarine and coastal waters during spring when phytoplankton dominated the plankton.

There has been considerable interest in recent years regarding the roles of microorganisms in marine ecosystems (see Pom- eroy 1974; Williams 198 1; Ducklow 1983) and much effort has been concentrated on measuring growth and productivity. In par- ticular, development of the frequency-of- dividing-cells (FDC) method (Hagstrom et al. 1979) and the thymidine-incorporation method (Fuhrman and Azam 1980, 1982) allowed a demonstration that heterotrophic bacterioplankton can consume lo-50% of the total primary production in coastal surface waters. Along similar lines, Karl (1979) introduced a method for measuring growth of microorganisms (not just bacteria) based on incorporation of tritiated adenine into RNA. Fuhrman and Azam (1980) and Mo- riarty (1986) pointed out that measuring DNA synthesis was more appropriate than RNA synthesis for estimating growth. Karl (198 1) modified the adenine method to add measurement of DNA as well as RNA synthesis. Recent publications have used only the DNA synthesis estimate to calculate production in units of carbon for water column and benthic environments (Karl and Knauer 1984; Karl et al. 1984; Winn and Karl 1984a; Burns et al. 1984; Craven and Karl 1984).

As distinct from the FDC and thymidine- incorporation methods, which target the nonphotosynthetic bacteria (Azam and Fuhrman 1984), the adenine method is de- signed to measure the production of all microorganisms, including heterotrophic bacteria and protozoa as well as autotrophic algae and chemolithotrophic bacteria (Burns et al. 1984; Karl et al. 1984; Karl and Winn 1984). All these organisms must be impli- cated because an integral part of the method is measurement of the specific activity of the extracted ATP pool, and ATP occurs in all organisms (Holm-Hansen and Booth 1966). Thus, the only organisms that do not contribute to the measurement are those that are rare or excluded by prescreening the sample. Many workers are now interested in interactions between groups of micro-

organisms, and because this method lumps them all together, it doesn’t allow resolution of these processes.

The conceptual basis for calculating production from adenine incorporation is described in detail by Karl et al. (198 la) and Winn and Karl (1984a) and will not be re- peated here. Exogenous tritiated adenine is assumed to be taken up and then converted to dATP before incorporation into DNA by all microorganisms. The rate of DNA synthesis is calculated and converted to carbon production via the C : DNA ratio. The equa- tion (broken into three parts) is

P = SA S = I/R R = E3H-ATP]/[ATP]

(1)

where P is the production in units of carbon (mass/volume/time), S is the DNA synthesis rate (mass/volume/time), A is the C : DNA ratio (mass/mass), 1 is the rate of 3H incorporation into DNA (radioactivity/ volume/time), and R is the ATP pool specific activity (radioactivity/mass). All the terms are multiplicative, as would be any errors.

A central assumption of the method is that all the organisms involved in any portion of the measurement must have a uniform response. As stated by Karl et al. ( 19 8 1 a, p. 2) under conditions for ecological application,

(a) All (or most) microorganisms assimilate adenine by a common and predictable pathway. (b) A uniform metabolic response is required if the technique is to be used for estimating community growth.

This assumption is required because the mixed microbial community is treated mathematically as if it is one composite or- ganism with one intracellular specific activity and one C : DNA ratio.

The reason for the importance of this assumption is clear if one considers that calculation of production (Eq. 1) is highly sen- sitive to variations in specific activity between organisms in the same sample; in this situation, it is not legitimate to use the average specific activity, which is what is measured. As an illustrative example, if some organisms do not take up adenine,

Adenine and microbjal production 629

they contribute only to the ATP concentration term (through their ATP biomass), but not to the [3H]ATP or incorporation terms. Thus, their biomass but not production fac- tors into the calculation of “total microbial production.” In the method, the total ATP of all the microorganisms is extracted together, so it is not possible to isolate the ATP from only those organisms assimilat- ing adenine.

Past literature on dissolved organic compounds other than adenine indicates that most or all of the uptake in natural seawater is by bacteria and not by algae or protozoa (e.g. Munro and Brock 1968; Williams 1970; Kirchman and Hodson 1984; Kirchman et al. 1985; Li and Dickie 1985). It is not clear why adenine should be expected to be any different in this regard. Low or nonexistent uptake by eucaryotes would lead to overestimates of S in the model described above.

Because the extent of adenine uptake by eucaryotes in field samples has not been well studied and because this process is of central importance to the adenine method, we examined by size fractionation, microauto- radiography, and metabolic inhibition whether eucaryotic organisms in natural samples (primarily phytoplankton, but also protozoa) were capable of significant adenine uptake compared to bacteria. Our results suggest they are not. We also point out that even if in selected environments eucaryotes managed to compete equally with bacteria for adenine, the C : DNA ratio of 50 used in reports to date is too high to apply to the bacteria, and variations in this ratio could make interpretation of adenine data difficult if not impossible.

We thank R. Hodson and F. Azam for discussions, D. Comeau, A. Hagstrdm, and C. Lee for reviewing the manuscript, and T. Bell, S. Hill, S. Pike, and A. Ryan for tech- nical assistance.

Materials and methods Samples were collected in acid-washed

plastic bottles from Crane Neck, on Long Island Sound, New York (40’5 5.3’N, 73”09.3’W; salinity about 28o/oo), and Horn Point Dock, on the Choptank River, Mary- land (38”35.0’N, 76’06.O’W; salinity about 7?~), a subestuary of Chesapeake Bay. Ex-

periments were done shortly after collec- tion, with samples kept under in situ or simulated in situ temperature and light conditions.

Size fractionation of [2-3H]adenine, [methyl-3H]thymidine, and [14C]sodium bicarbonate uptake was done with various pore-size 25-mm-diam Nuclepore filters (Azam and Hodson 1977). Samples were incubated in sterile Whirlpak bags or plastic centrifuge tubes. Labeled adenine (33 Ci mmol-l, Research Products Int.) and thymidine (70 Ci mmol-l, New England Nu- clear) were added at 1-5 nM concentrations (equimolar within experiments); bicarbonate (0.1 PCi hg-I, New England Nuclear) was added at 1.2 PM. Incubations lasted 0.5-l .5 h (2.5-3 h for 14C) and were ended by fil- tration of lo-30 ml (50 ml for 14C) subsamples. Filters for 3H uptake were rinsed five times with filtered seawater and placed in scintillation vials. The 14C filters were treated with 30 ~1 of glacial acetic acid to elim- inate inorganic 14C. Radioactivity was counted by liquid scintillation and correct- ed for quench by the external standard method. All samples were analyzed in duplicate and compared to 1% Formalin-killed (for 3H) or zero-time (for 14C) controls.

In parallel to the uptake measurements, chlorophyll and bacterial direct counts were also size-fractionated. Duplicate 20-50-ml samples were filtered through various pore- size 25-mm-diam Nuclepore filters. The filters were extracted with 90% acetone at -20°C for 24 h and fluorometrically analyzed for chlorophyll a (Parsons et al. 1984). The filtrates were preserved with 5% borate- buffered Formalin and bacteria counted by acridine orange epifluorescence direct counts on 0.2-pm pore-size Nuclepore filters (Hob- bie et al. 1977) on an Olympus BH-2 or Nikon Optiphot microscope outfitted with HBO-100 mercury vapor lamps. All anal- yses were done in duplicate.

Data from size fractionations are pre- sented as the percentage caught on a given filter of that on the 0.2-pm pore-size filter. The “percentages caught” of bacterial counts, which were from filtrates, were calculated as [ 1 - (filtrate/unfiltered)] x 100.

Production by heterotrophic bacteria was calculated from thymidine incorporation

630 Fuhrman et al.

Table 1. Bacterial and phytoplankton biomass estimates.

Bact.* Phyt0.t Bact./ phyto.

Location 1985 (pg C liter-l) (W

Long Island 14 Feb 8.6 276 3 Long Island 21 Feb 11.6 376 3 Long Island 14 Mar 27.9 1,696 2 Long Island 9 Apr 46.1 438 11 Long Island 12 Apr 34.1 320 11 Choptank 24 Apr 96.0 2,520 4

* Direct cell counts x 10 fg C cell-‘. t Chlorophyll x 40.

into cold TCA-insoluble material with a conversion factor of 2 x 10 1 8 cells produced per mole thymidine incorporated (Fuhrman and Azam 1982; Kirchman et al. 1982; Bell and Kuparinen 1984). Cell biomass was not measured in the individual samples of this study, but previous measurements in Long Island Sound and Chesapeake Bay from epifluorescence photomicrographs indicate an average bacterial carbon content of about 10 fg, similar to that for southern California (Fuhrman 198 1).

Autoradiography was the MARG-E pro- cedure described by Tabor and Neihof (1982), with thymidine- and adenine-labeled samples (1 nM additions) from Long Island Sound (14 March). Exposure time was 3 days. Controls for background counts were killed with 1% Formalin. We counted the individual silver grains associated with phytoplankton and bacteria, including only those grains within 1 pm of a recognizable cell. Grains within 1 pm of both bacterial and algal cells were counted as algal (dis- cussed below).

Adenine uptake by freshly collected ciliates was measured by picking out oligotrich ciliates (mostly tintinnids, average length about 50 pm) from a 20-pm net tow (10 April 198 5 at Crane Neck). Sixty ciliates were suspended in 10 ml of 0.22~pm filtered seawater. A control sample had 10 ml of unfiltered seawater. [3H]adenine (6 nM) was added to both samples, and they were incubated under simulated in situ conditions for 1 h. Midway through the incubation, the sample was examined microscopically to verify that the ciliates were active and swimming. Duplicate 5 -ml subsamples were serially filtered through 3-pm and 0.2-pm

pore-size Nuclepore filters, which were then rinsed five times with filtered seawater and radioassayed by liquid scintillation.

Inhibition of adenine, thymidine, and [ 14C]bicarbonate uptake by cycloheximide was measured from Long Island Sound samples on 2 1 May. Of six 1 OO-ml samples, three were pretreated for 1 h at simulated in situ light and temperature by addition of 0.5 ml of a 10 mg ml-’ aqueous solution of cycloheximide (Sigma: final concn, 50 mg liter-‘). The three controls had 0.5 ml of distilled water added. To one treated and one control sample each was added 5 nM [2-3H]adenine, or 3.5 nM [methyl- 3H]thymidine, o r 1.4 PM [14C]bicarbonate. Incubation was under simulated in situ conditions and lo-ml subsamples were filtered four times over a 1.5-h time-course through 25-mm-diam Millipore filters of 0.45-pm nominal pore-size. Filters were rinsed three times with about 1 ml of filtered seawater, and radioactivity was measured by liquid scintillation.

Results The size fractionation and autoradiogra-

phy experiments were done under spring bloom conditions, when phytoplankton greatly dominated both biomass and production. These conditions were ideal for studies of size fractionation because almost all the phytoplankton were >8 pm and overlapped minimally in size with bacteria. Under these conditions, with active large phytoplankton as the dominant biomass component, one would expect to see uptake of adenine in larger size fractions if the assumptions of the adenine method were ful- filled. Biomass estimates from chlorophyll (C : Chl ratio of 40: Burns et al. 1984) and bacterial cell counts indicate the extent of phytoplankton domination in terms of total cell carbon (Table 1). Productivity showed a comparable domination, with bacterial heterotrophic production on 14 March (peak of bloom) of 0.5 pg C liter-’ h-l compared to 79 pg C liter-l h-l primary production (bacterial/total = 0.6%); on 12 April the rates were 0.4 pg C liter-’ h-l bacterial and 12 pg C liter-’ h-l primary production (bacterial/ total = 3.3%). In the Choptank River sam-

Adenine and microbial production 631

NUCLEPORE FILTER PORE SIZE IN pm

m Thymidine Uptake a Bacterial Abundance a Adenine Uptake

m Chlorophyl I Q Concentration m 14C Bicarbonate Uptake

Fig. 1. Size distribution of measured parameters. A-E-Long Island Sound; F-Choptank River. Sampling dates: A- 14 February; B-2 1 February; C- 14 March; D- 9 April; E- 12 April; F-24 April.

ples, the bacterial production was 5.7 pg C true in the Choptank River. In any case, the liter-l h-l and primary production was 44 fraction of total uptake in the large size frac- hg C liter-’ h-’ (bacterial/total = 13%). tions was still very small.

Still, even in these phytoplankton-dominated systems, the adenine uptake was dominated by the smallest organisms, as was thymidine uptake. The size distributions for both processes were very similar to the distribution of bacterial abundance and con- trasted sharply with the distribution of chlorophyll and primary production (Fig. 1). In the March and April Long Island Sound samples, there was evidence that some large size fractions took up proportionately more adenine than thymidine; the opposite was

The observation that the adenine and thymidine uptake size fractionation generally followed that of bacterial cells, almost all of which were < 1 pm, suggests that most if not all of the uptake in the large sizes was due to attached bacteria, not to large organisms. The 14 February sample had a relatively large percentage of adenine and thymidine uptake and bacterial counts in the large size fractions. This is consistent with the observation that considerable detritus was visible in this sample, perhaps from

632 Fuhrman et al.

Table 2. Autoradiography of adenine and thymidine uptake.

Cell type Substrate* Cells liter-l? With grains+ Avg grains per Total grains*

No. examined (W cell* (W

Bacteria A 5.1 x 109 345 21 0.25 97.6 Bacteria T 4.6 x lo9 315 33 0.48 97.5 Phytoplankton A 3.2 x lo6 100 92 10.1 2.4 Phytoplankton T 4.1 x106 100 94 13.8 2.5

* A-adenine; T-thymidine. j’ Determined from autoradiography slide. $ Phytoplankton values are overestimates due to inclusion of grains associated with coincident bacteria (see text).

decaying macroalgae or seagrass suspended during a recent storm.

The size fractionation data do not rule out the possibility that the phytoplankton took up some portion of the adenine, albeit a small one. We used autoradiography, which is only semiquantitative, to look for label directly associated with phytoplankton. In general, the phytoplankton were not heavily labeled with either thymidine or adenine. It appeared that most if not all of the silver grains near the phytoplankton were probably associated with bacteria that were on or under the large phytoplankton cells, as was found for glutamate by Brock and Brock (1966) and thymidine by Bern (1985). However, we counted these grains as associated with the phytoplankton anyway, to be certain that we were not missing any algal uptake. Therefore, the phytoplankton grain counts are probably large overestimates, and these indicate that <2.5% of the uptake was by phytoplankton (Table 2).

There was no evidence of any significant amount of adenine uptake by the ciliates. Ciliates were concentrated about 14 times above natural abundance levels, yet the radioactivity in ciliates collected on the 3-pm filter was only 59 dpm above background; the unfiltered seawater uptake was 2 1,324 dpm >0.2 pm and 366 dpm >3 pm. The enhanced ciliate biomass was calculated to be 9.6 hg C liter-‘, compared with 34 pg C liter-l bacterial biomass in the control.

Cycloheximide, a eucaryote-specific in- hibitor (Ennis and Lubin 1964), substan- tially inhibited the assimilation of [14C]bicarbonate, but not of thymidine or adenine; the assimilation rates of treated samples as percent of control samples were 9 1% for adenine, 88% for thymidine, and 38% for bicarbonate; in other words, < 10%

of the adenine uptake was affected. The inhibition of bicarbonate assimilation suggests that most of the carbon fixation was by eucaryotes, in accordance with the observation that at this time of year, few if any cyanobacteria were present (L. Campbell pers. comm.). The much smaller effect on thymidine and adenine assimilation is fur- ther evidence that these substrates were uti- lized almost exclusively by bacteria.

Discussion Adenine assimilation -Our data show

convincingly that different groups of microorganisms in field samples have highly nonuniform responses with respect to adenine assimilation. The size fractionation and autoradiography results show that bacteria outcompeted phytoplankton for adenine uptake even though phytoplankton dominated both in terms of total biomass and total production. This was true for all size- fractionation samples; about 80-95% of the adenine uptake was < 8 pm (along with bacterial abundance and thymidine uptake), while 65-l 00% of the chlorophyll and bicarbonate uptake was >8 pm. More than 97% of the silver grains in the adenine autoradiography were associated with bacteria. The experiments with ciliates showed that there was no indication of adenine uptake by these organisms even when their biomass was about a third that of the bacteria. Finally, the cycloheximide inhibition experiment confirmed that adenine utilization was predominantly procaryotic.

A simple calculation from the productivity, size fractionation, and autoradiography data shows that there was not a uniform adenine metabolism by bacteria and eucaryotes, as required by the adenine meth-


od. Given the disparity in uptake rates, biomass, and productivities, the adenine taken up by bacteria would have had to contribute << 1% to their dATP pools, while that taken up by phytoplankton, if there was any, would simultaneously have had to contribute > 95% for there to be similar dATP specific activities. This would in itself show a highly nonuniform response.

Karl et al. (198 1 b) reported one size-fractionation measurement in marine waters at a coastal station. Our recalculation of their results indicates that the < 12-pm size fraction incorporated radioactivity (PCi liter-l h-l) 49% and 54% as fast as unfractionated water, with preincubation and postincuba- tion size fractionation, respectively. The specific activities of the two replicate unfractionated samples, however, varied by more than a factor of two, so the low pre- cision of this measurement of nucleic acid synthesis makes comparison of these rates difficult. Also, because their “total” uptake was collected on Reeve-Angel glass-fiber filters, which allow about 50% of the bacteria to pass through (Fuhrman and Azam 1980), the percentage in the small size fraction is an underestimate. These filters, or the similar Whatman type GF/F, have been used in all previous adenine studies-possibly missing considerable bacterial uptake, These results do not necessarily demonstrate adenine utilization by eucaryotes because there is no indication of the abundance of attached bacteria in this sample. A large number of attached bacteria, as Karl et al. (198 1 b) reported for their freshwater sampling site and as we observed to a lesser extent in some samples (Fig. 1A; cf. Ducklow 1982), could explain the observation. It is important in this sort of experiment to back up the rate measurements with size fractionation of bacterial counts (free and attached) and chlorophyll.

We are not claiming that algae cannot take up adenine. Some species of phytoplankton may be able to do so (Winn and Karl 1984b), although to our knowledge this has not yet been shown in axenic cultures, except for one strain of cyanobacteria (Cuhel and Wa- terbury 1984). However, adenine uptake in a culture may not predict the behavior of diverse mixed assemblages in nature. The important question is not whether phyto-

plankton can take up adenine in culture, but rather whether phytoplankton and protozoa can compete effectively with natural bacteria for adenine uptake and incorporation under field conditions. In our experiments eucaryotes took up virtually no adenine even when their biomass and productivity over- whelmed those of bacteria. The concentrations of adenine that we added (l-6 nM) are comparable to those used in previous field studies (4-7 nM: e.g. Karl 1982; Winn and Karl 1984a; Karl and Knauer 1984; Burns et al. 1984).

The bacterial domination of adenine uptake would have important consequences for the interpretation of adenine incorporation data. As we pointed out above, the method requires a uniform response by all the microorganisms. Lack of adenine uptake by some or all eucaryotes would mean that eucaryotic ATP pools can be virtually unlabeled, so that the incorporation data can have little to do with their nucleic acid synthesis. However, the specific activity term would still be strongly influenced by eucaryotic ATP, which usually dominates (Williams 198 1; Holligan et al. 1984). Un- der these conditions, the biomass, but not growth, of the eucaryotes influences the result. This can make the method inappro- priate for measuring production rates in surface waters and other habitats where eucaryotic biomass dominates, such as sinking detritus (Silver et al. 1984).

It is an interesting consequence of Eq. 1 that if the nucleic acid synthesis rates in eucaryotes and bacteria are exactly propor- tional to the total ATP contents of these groups (Or [Seucaryo*es/Sbacteria 1 = [ATP,,,,,,,,,,/ATP,,.,,,,1) then the total nucleic acid synthesis rate calculated by the adenine method can be correct by offsetting errors if all the uptake is bacterial. This observation means that in natural samples where this condition may be approximately met, the adenine method could provide total nucleic acid synthesis rates that are of the right order of magnitude, appearing rea- sonable. However, the accuracy would de- pend completely on an exact proportion- ality between eucaryotic ATP-normalized nucleic acid synthesis and that of bacteria. There is no reason to expect this condition to be satisfied in nature.

634 Fuhrman et al.

DNA content-variability in the DNA content of different organisms will also af- fect the calculation of production by adenine incorporation. From Eq. 1, accurate knowledge of the C : DNA ratio is as important as any other variable for calculating production from adenine incorporation. Karl and Knauer (1984), Karl et al. (1984), Winn and Karl (1984a), Burns et al. (1984), and Craven and Karl (1984) use a C : DNA ratio of 50, derived from a literature report by Holm-Hansen (1969). But in that report, this ratio refers specifically to algae. We suggest here that this number may be an order of magnitude too high when applied to marine bacteria, which can dominate adenine uptake.

Literature reports show that a C : DNA ratio of 50 is not appropriate for all marine microorganisms, particularly bacteria. For algal cultures alone, the ratio ranges from 40 to 200 (Holm-Hansen et al. 1968). The DNA content of natural bacteria appears to be much higher. Fuhrman and Azam ( 19 82) measured the DNA in the 0.2-0.6~pm (bacterial) size fraction of seawater and con- cluded that most natural bacteria have C : DNA ratios near 5. Paul and Carlson (1984) measured DNA and particulate C in size fractions of natural waters and found C : DNA ratios as low as 2.3 (avg 4) in the bacterial size fraction (0.2-l gm). They con- cluded that most of the particulate DNA in the marine environment is associated with bacterioplankton and not phytoplankton. They also considered and rejected the possibility that a significant portion of the DNA in the bacterial size fraction was detrital, in agreement with the conclusion of Dortch et al. (1983).

Fuhrman and Azam (1982) have argued that the small size of natural marine bacteria must result in a relatively high DNA content; they calculated that at a C : DNA ratio of 33, many small bacteria would have enough DNA for only about 50 genes, far too few for an independently living organ- ism. Even though marine bacteria are small, they still need the genetic information to cope with a complex environment. The measurements of Fuhrman and Azam (1982) and Paul and Carlson (1984) indicate a genome size for marine bacteria approx-

imately equal to that of previously studied eubacteria, as may be expected. Most bacteria cultured in nutrient broth have typi- cally five times the diameter or 125 times the volume of natural marine bacteria, so that with the same genome size, they have a much larger C : DNA ratio, inapplicable for field studies.

Therefore, even if one could find an environment where adenine uptake and assimilation are uniform among microorganisms, the very large range of C : DNA ratios (N 4-200) among microorganisms makes it difficult to choose an appropriate ratio for calculating production from adenine incorporation data. In any event, a ratio of 50 seems too high to apply to natural marine bacteria. Thus, the current adenine-based production estimates are probably several- fold high, particularly when bacterial production is important. The extent of overestimation could vary significantly, de- pending on the relative abundance and activity of different microorganisms.

Given that the algal domination of most waters of the euphotic zone brings the adenine results from that habitat into question, one might ask how well the method works in other environments. For example, Karl and Knauer (1984) reported on production inside sediment traps. But Silver et al. (1984) reported that in the same sediment traps, microscopic examination showed that ciliated protozoa could sometimes account for 90% of the biomass. The lack of demonstrable adenine incorporation by ciliates thus calls into question the microbial production results for these sediment traps. Also, Winn and Karl (1984a) worked both within and below the photic zone and asserted that below the photic zone nucleic acid metabolism is predominantly due to heterotrophic bacteria. However protozoa and other eucaryotes, such as sinking viable phytoplankton (Smayda 197 1; Platt et al. 1983), may be important there and need to be considered in studies of this type. As a final note, we suggest that automatic interpretation of adenine incorporation in terms of growth may be premature for organisms such as chemolithotrophs (Karl et al. 1984) that have not yet been specifically studied in this regard.


References AZAM, F., AND J. A. FUHRMAN. 1984. Measurement

of bacterioplankton growth in the sea and its reg- ulation by environmental conditions, p. 179-l 96. In J. E. Hobbie and P. J. Williams [cds.], Hetero- trophic activity in the sea. Plenum.

-, AND R. E. HODSON. 1977. Size distribution and activity of marine microheterotrophs. Lim- nol. Oceanogr. 22: 492-50 1.

BELL, R. T., AND J. KUPARINEN. 1984. Assessing phytoplankton and bacterioplankton production during early spring in Lake Erken, Sweden. Appl. En- viron. Microbial. 48: 122 l-l 230.

BERN, L. 1985. Autoradiographic studies of [methyl- 3H] thymidine incorporation in a cyanobacterium (Microcystis wesenbergii)-bacterium association and in select algae and bacteria. Appl. Environ. Microbial. 49: 232-233.

BROCK, T. D., AND M. L. BROCK. 1966. Autoradiog- raphy as a tool in microbial ecology. Nature 209: 734-736.

BURNS, D., AND OTHERS. 1984. Microbial biomass, rates of DNA synthesis and estimated carbon production in Kaneohe Bay, Hawaii. Bull. Mar. Sci. 34: 346-357.

CRAVEN, D. B., AND D. M. KARL. 1984. Microbial RNA and DNA synthesis in marine sediments. Mar. Biol. 83: 129-139.

CUHEL, R. L., AND J. B. WATERBURY. 1984. Bio- chemical composition and short term nutrient incorporation patterns in a unicellular marine cyanobacterium, Synechococcus (WH7803). Limnol. Oceanogr. 29: 370-374.

DORTCH, Q., T. L. ROBERTS, J. R. CLAYTON, JR., AND S. I. AHMED. 1983. RNA/DNA ratios and DNA concentrations as indicators of growth rate and biomass in planktonic marine organisms. Mar. Ecol. Prog. Ser. 13: 61-71.

DUCKLOW, H. W. 1982. Chesapeake Bay nutrient and plankton dynamics. 1. Bacterial biomass and production during spring tidal destratification in the York River, Virginia, estuary. Limnol. Oceanogr. 27: 65 l-659.

- 1983. Production and fate of bacteria in the oceans. Bioscience 33: 494-50 I.

ENNIS, H. L., AND M. LUBIN. 1964. Cycloheximide: Aspects of inhibition of protein synthesis in mam- malian cells. Science 146: 1474-1476.

FUHRMAN, J. A. 198 1. Influence of method on the apparent size distribution of bacterioplankton cells: Epifluorescence microscopy compared to scanning electron microscopy. Mar. Ecol. Prog. Ser. 5: 103- 106.

AND F. AZAM. 1980. Bacterioplankton sec- ondary production estimates for coastal waters of British Columbia, Antarctica, and California. Appl. Environ. Mirobiol. 39: 1085-1095.

-, AND -. 1982. Thymidine incorporation as a measure of heterotrophic bacterioplankton production in marine surface waters: Evaluation and field results. Mar. Biol. 66: 109-I 20.

HAGSTR~M, A., U. LARSSON, P. HORSTEDT, AND S. NORMARK. 1979. Frequency of dividing cells, a new approach to the determination of bacterial

growth rates in aquatic environments. Appl. En- viron. Microbial. 37: 805-8 12.

HOBBIE, J. E., R. J. DALEY, AND S. JASPER. 1977. Use of nuclepore filters for counting bacteria by fluorescence microscopy. Appl. Environ. Microbial. 33: 1225-1228.

HOLLIGAN, P. M., AND OTHERS. 1984. Vertical distribution and partitioning of organic carbon in mixed, frontal, and stratified waters of the English Channel. Mar. Ecol. Prog. Ser. 14: 111-127.

HOLM-HANSEN, 0. 1969. Algae: Amounts of DNA and organic carbon in single cells. Science 163: 87-88.

-, AND C. R. BOOTH. 1966. The measurement of adenosine triphosphate in the ocean and its ecological significance. Limnol. Oceanogr. 11: 5 1 O- 519.

-, W. H. SUTCLIFFE, JR., AND J. SHARP. 1968. Measurement of deoxyribonucleic acid in the ocean and its ecological significance. Limnol. Oceanogr. 13: 507-5 14.

KARL, D. M. 1979. Measurement of microbial activity and growth in the ocean by rates of stable ribonucleic acid synthesis. Appl. Environ. Microbial. 38: 850-860.

- . 198 1. Simultaneous rates of ribonucleic acid and deoxyribonucleic acid syntheses for estimating growth and cell division of aquatic microbial communities. Appl. Environ. Microbial. 42: 802- 810.

- 1982. Selected nucleic acid precursors in . studies of aquatic microbial ecology. Appl. En- viron. Microbial. 44: 89 l-902.

-, AND G. A. KNAUER. 1984. Vertical distribution, transport, and exchange of carbon in the northeast Pacific Ocean: Evidence for multiple zones of biological activity. Deep-Sea Res. 31: 22 l- 243.

- -, J. H. MARTIN, AND B. B. WARD. 1984. Bacterial chemolithotrophy in the ocean is associated with sinking particles. Nature 309: 54-56,

-, AND C. D. WINN. 1984. Adenine metabolism and nucleic acid synthesis: Applications to micro- biological oceanography, p. 197-2 16. In J. E. Hob- bie and P. J. Williams [eds.], Heterotrophic activity in the sea. Plenum.

- - synthesis as

AND D. C. WONG. 198 la,b. RNA a measure of microbial growth in

aquatic environments. 1. Evaluation, verification, and optimization of methods. 2. Field applications. Mar. Biol. 64: I-12, 13-2 1.

KIRCHMAN, D., H. W. DUCKLOW, AND R. MITCHELL. 1982. Estimates of bacterial growth from changes in uptake rates and biomass. Appl. Environ. Mi- crobiol. 44: 1296-l 307.

-, AND R. HODSON. 1984. Inhibition by pep- tides of amino acid uptake by bacterial popula- tions in natural waters: Implications for the reg- ulation ofamino acid transport and incorporation. Appl. Environ. Microbial. 47: 624-63 1.

-, E. K’NEES, AND R. HODSON. 1985. Leucine incorporation and its potential as a measure of protein synthesis by bacteria in natural aquatic systems. Appl. Environ. Microbial. 49: 599-607.

636 Fuhrman et al.

LI, W. K., AND P. M. DICKIE. 1985. Metabolic inhibition of size-fractionated marine plankton ra- diolabeled with amino acids, glucose, bicarbonate, and phosphate in the light and dark. Microb. Ecol. 11: 11-24.

MORIARTY, D. J. 1986. Measurement of bacterial growth rates in aquatic systems from rates of nucleic acid synthesis. Adv. Microb. Ecol. 9 (in press).

MUNRO, A. L., AND T. D. BROCK. 1968. Distinction between bacterial and algal utilization of soluble substances in the sea. J. Gen. Microbial. 51: 35- 42.

PARSONS, T. R., Y. MAITA, AND C. M. LALLI. 1984. A manual of chemical and biological methods for seawater analysis. Pergamon.

PAUL, J. H., AND D. J. CARLSON. 1984. Genetic material in the marine environment: Implication for bacterial DNA. Limnol. Oceanogr. 29: 109 l-1097.

PLATT, T., AND OTHERS. 1983. Photosynthetically- competent phytoplankton from the aphotic zone of the deep ocean. Mar. Ecol. Prog. Ser. 10: 105- 110.

POMEROY, L. R. 1974. The ocean’s food web; a chang- ing paradigm. Bioscience 24: 499-504.

SILVER, M.W.,M.M. GOWING, D.C. BROWNLEE, AND J. 0. CORLISS. 1984. Ciliated protozoa associated with oceanic sinking detritus. Nature 309: 246- 248.

SMAYDA, T. J. 197 1. Normal and accelerated sinking of phytoplankton in the sea. Mar. Geol. 11: 105- 122.

TABOR, P. S., AND R. A. NEIHOF. 1982. Improved microautoradiographic method to determine individual miroorganisms active in substrate uptake in natural waters. Appl. Environ. Microbial. 44: 945-953.

WILLIAMS, P. J. 1970. Heterotrophic utilization of dissolved organic compounds in the sea. I. Size distribution of population and relationship between respiration and incorporation of growth substrates. J. Mar. Biol. Assoc. U.K. 50: 859-870.

-. 198 1. Incorporation of microheterotrophic processes into the classic paradigm of the planktonic food web. Kiel. Meeresforsch. Sonderh. 5: l-28.

WINN, C. D., AND D. M. KARL. 1984a. Microbial productivity and community growth rate estimates in the tropical north Pacific Ocean. Biol. Oceanogr. 3: 123-145.

-, AND -. 1984b. Laboratory calibration of the t3H]adenine technique for measuring rates of RNA and DNA synthesis in marine microorganisms. Appl. Environ. Microbial. 47: 8 3 5-842.

Submitted: 7 June 1985 Accepted: 6 December 1985

Notice

It has been pointed out by N. Hojerslev that the Secchi disk depth record of 5 3 m claimed for the eastern Mediterranean by Berman et al. ( 198 5) was exceeded by an observation of a transparency of 66.5 m in the Sargasso Sea, made by lowering a 2-m square of sailcloth into the water (Kriimmel 1907). Dr. Hojerslev also cites other records, somewhat less well documented.

Any additional communications on this subject should be addressed to the Guinness Book of World Records, not to LIMNOLOGY AND OCEANOGRAPHY.

Yvette H. Edmondson Editor

References

BERMAN, T., P. D. WALLINE, A. SCHELLER, J. ROTHENBERG, AND D. W. TOWNSEND. 1985. Secchi disk depth record: A claim for the eastern Mediterranean. Limnol. Oceanogr. 30: 447-448.

KRGMMEL, 0. 1907. Handbuch der Ozeanographie. V. 1. Englehorn.

does adenine incorporation into nucleic acids measure total microbial production?

Documents