APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Aug. 1988, p. 1934-1939 Vol. 54, No. 80099-2240/88/081934-06$02.00/0Copyright © 1988, American Society for Microbiology

Estimating Bacterial Production in Marine Waters from theSimultaneous Incorporation of Thymidine and Leucine

GERARDO CHIN-LEO* AND DAVID L. KIRCHMAN

College of Marine Studies, University of Delaware, Lewes, Delaware 19958

Received 8 February 1988/Accepted 3 May 1988

We examined the simultaneous incorporation of [3H]thymidine and ['4C]leucine to obtain two independentindices of bacterial production (DNA and protein syntheses) in a single incubation. Incorporation rates ofleucine estimated by the dual-label method were generally higher than those obtained by the single-labelmethod, but the differences were small (dual/single = 1.1 + 0.2 [mean ± standard deviation]) and wereprobably due to the presence of labeled leucyl-tRNA in the cold trichloroacetic acid-insoluble fraction. Therewere no significant differences in thymidine incorporation between dual- and single-label incubations (dual/single = 1.03 ± 0.13). Addition of the two substrates in relatively large amounts (25 nM) did not apparentlyincrease bacterial activity during short incubations (<5 h). With the dual-label method we found thatthymidine and leucine incorporation rates covaried over depth profiles of the Chesapeake Bay. Estimates ofbacterial production based on thymidine and leucine differed by less than 25%. Although the need forappropriate conversion factors has not been eliminated, the dual-label approach can be used to examine thevariation in bacterial production while ensuring that the observed variation in incorporation rates is due to realchanges in bacterial production rather than changes in conversion factors or introduction of other artifacts.

Estimates of the rate of biomass production have beeninstrumental in determinations of the relative importance ofheterotrophic bacteria as biomass producers and as miner-alizers of organic matter in aquatic ecosystems. The amountof bacterial biomass potentially available to grazers and thusto higher trophic levels can be determined from rates ofbacterial production. Coupled with information about assim-ilation efficiency, production rates can also be used toestimate the total uptake of organic matter by bacteria. Inaddition, rates of bacterial production can be used to esti-mate the average growth rate of bacterial assemblages and asan indicator of the response of bacteria to fluctuations inenvironmental conditions.Of the various methods available to measure bacterial

production, [3H]thymidine incorporation into cold trichloro-acetic acid (TCA)-insoluble material (5-7, 20) has becomethe most widely used. However, the conversion of thymi-dine incorporation rates into reliable production estimatescan be problematic. To achieve this conversion, a factor isoften applied that is derived either theoretically from as-sumptions of the extent of isotope dilution, the amount ofDNA per cell, and the thymine content of bacterial DNA (6)or empirically by comparing incorporation rates with in-creases in bacterial numbers under controlled conditions (15,21). Since the empirical factor, as well as rates of thymidineincorporation, may vary (1, 4, 15, 21), it is unclear whetherchanges in the calculated rate of production are real or aredue to variations in the conversion factor. Although varia-tions in the conversion factor can be determined empirically,the required experiments are time-consuming, and it isimpractical to perform them concurrently with incorporationmeasurements. A way to circumvent this limitation is tocompare rates of thymidine incorporation with other mea-surements of bacterial production. Agreement between in-dependent measurements made simultaneously providesconfidence that observed variations reflect real changes inrates of bacterial production.

* Corresponding author.

Leucine incorporation into protein (hot-TCA-insolublematerial) has been proposed as a measure of protein synthe-sis and bacterial biomass production (14, 17). Kirchman etal. (17) demonstrated that increases in leucine incorporationagreed with increases in cell numbers and protein content.Proteins account for a large percentage of bacterial biomass(ca. 50%) in pure culture (13) and in natural assemblages (9),and their synthesis consumes a large percentage of thecellular energy. Therefore, the rate of protein synthesis maybe a good indicator of total biomass production and energyutilization. Rates of leucine incorporation could provide anindependent check of the thymidine method. McDonough etal. (18) found that [3H]leucine incorporation covaried with[3H]thymidine incorporation into protein but not with[3H]thymidine incorporation into DNA in depth profiles of alake with an anoxic hypolimnion. More comparisons of thetwo methods are needed.

In this study we used a dual-label radioisotope method tomeasure leucine and thymidine incorporation in a singleincubation. This approach simplifies the simultaneous mea-surement of these incorporation rates, reduces the cost inlabor and materials, and minimizes errors associated withrepeated subsampling. We found that rates of bacterialproduction as estimated by thymidine and leucine incorpo-ration covaried and differed by less than 25%.

MATERIALS AND METHODS

Sampling sites. Water samples were collected from theChesapeake Bay, Delaware Bay, Roosevelt Inlet (Lewes,Del.), and the Mid Atlantic Bight during the spring andsummer of 1986 and 1987. Exact locations are given with theresults of each experiment. The Chesapeake Bay, DelawareBay, and Mid Atlantic Bight samples were obtained fromvarious depths with 10 liter Nisken bottles by using aNeil-Brown automated rosette sampler aboard the R.V.Cape Henlopen or the R.V. Cape Hatteras. Roosevelt Inletsamples were obtained from surface waters off the College ofMarine Studies dock. Experiments were performed immedi-

1934

on May 16, 2018 by guest

http://aem.asm

.org/D

ownloaded from

BACTERIAL PRODUCTION IN MARINE WATERS 1935

ately after collection, and samples were kept in the dark atthe in situ temperature.

Incorporation of [3H]thymidine and [3lH]leucine: single-label incubations. We examined the incorporation rate of[3H]thymidine into cold-TCA-insoluble material (6) and of[3H]leucine into hot-TCA-insoluble material (14). Duplicatesamples (10 ml) were incubated with either 5 nM (finalconcentration) [3H]thymidine (specific activity, 84.1 Cimmol-') or 11 nM (final concentration) leucine (1 nM[3H]leucine [specific activity, 60 Ci mmol-1], 10 nM nonra-dioactive leucine). Incubation times ranged from 30 to 60min. Abiotic adsorption or radioactivity was measured withTCA-killed controls. Incubations were ended by coolingsamples in an ice-cold water bath for 1 min and then adding1 ml of 50% TCA to reach a final concentration of 5%.[3H]thymidine samples were cooled for an additional 5 minto extract the macromolecular fraction. [3H]leucine sampleswere heated for 15 min at 85°C to hydrolyze all macromol-ecules except protein. Following extraction, samples werefiltered through Gelman filters (pore size, 0.45 pLm). Thefilters were rinsed twice with 1 ml of ice-cold 5% TCA andonce with distilled water and then radioassayed. All radio-active substrates were from New England Nuclear Corp.Nonradioactive substrates were from Sigma Chemical Co.

Incorporation of [3llthymidine and [4C]ileucine: dual-labelincubations. To measure simultaneously the incorporation ofthymidine and leucine, we added [3H]thymidine and[14C]ileucine to a single sample and collected the cold-TCA-insoluble material. [14C]ileucine was chosen because [2-14C]thymidine can undergo catabolism and still label DNAowing to the retention of the label in uracil (8). Duplicatewater samples (10 ml) were incubated with 5 nM (finalconcentration) [3H]thymidine (specific activity, 84.1 Cimmol-1) and 20 nM (final concentration) leucine (10 nM[14C]ileucine [specific activity, 328.5 mCi mmol-1], 10 nMnonradioactive leucine). Because of the low specific activityof [14C]ileucine, a higher final concentration (20 nM) wasused initially than that of [3H]leucine (10 nM). However, wehave been able to measure [14C]ileucine incorporation with a10 nM addition even in oligotrophic waters (D. L. Kirchman,unpublished data). Incubation times ranged from 30 to 60min. Termination of incubation, cold-TCA extraction ofmacromolecules, subsequent filtration, and preparation offilters for radioassay proceeded as described above.

Conversion factors. Conversion factors for thymidine andleucine incorporation in dual- and single-label procedureswere calculated by comparing increases in cell numbers withthe total amount of substrate incorporated (15). Water sam-ples (200 ml) were collected from various depths in the upperChesapeake Bay during May and July 1987. To minimizegrazing of bacteria by protozoa (22), we diluted samples in aratio of 1:9 with filtered (pore size, 0.22,um) seawater fromthe collection site. Samples (final volume, 2 liters) were thenkept in darkened polycarbonate bottles (Nalgene) and incu-bated at the surface seawater temperature. The bottles weresubsampled every 6 h for a total of 48 h to measure changesin bacterial abundance and the incorporation of thymidineand leucine during the dual- and single-label procedures.Bacterial abundance was measured by using acridine orangeand epifluorescence microscopy (12).Changes in bacterial abundance and total substrate incor-

poration were calculated for each sampling interval. Theconversion factor was defined as the ratio of total increasesin bacterial abundance to the total integrated incorporation.Errors for bacterial abundance and substrate incorporationwere propagated to estimate the uncertainty of the conver-

4

E3

0 2 Hot TCA

Time (min)FIG. 1. Incorporation of [3H]leucine into cold- and hot-TCA-

insoluble material over time. Samples were obtained in RooseveltInlet, Lewes, Del. Measurements were done in triplicate; standarderrors were less than 10%.

sion factor (2). Estimates of bacterial production derivedfrom these conversion factors were compared with estimatesobtained by Fuhrman and Azam (6) with conversion factorsfor thymidine.Measurement of incorporated radioactivity. Radioactivity

incorporated into cellular material was counted with a Beck-man LS 3801 liquid scintillation spectrometer. Quenchingwas corrected with the external standard in the Comptonedge shift mode (H#). Spillover of "'C into the 3H spectrumand vice versa were corrected with the Beckman AutomatedQuench Control program based on H#. Reading windowswere adjusted with the Automated Quench Control programas a function of quenching (H#) to minimize "'C spill. TheBeckman manual for this instrument warns against attempt-ing to calculate disintegrations per minute in a dual-labelexperiment in which the ratio of "'C to 3H counts per minuteis greater than 15. In our experiments, the "'C/3H ratio ofcounts per minute incorporated into cellular material wasnever greater than 1.

RESULTS

The dual-label method consists of adding both [3H]thymi-dine and ["'Cileucine to a single sample. After incubation thecold-TCA insoluble material is collected on filters. Bacterialproduction is estimated from incorporation rates of theradiolabeled compounds by using measured (see below)conversion factors or factors taken from the literature (6).Hot versus cold extraction of leucine incorporation. The

leucine method involves the use of a hot-TCA extraction tohydrolyze the nonprotein macromolecules. In the dual-labelapproach, leucine incorporation into leucyl-tRNA whichremains in the cold-TCA-insoluble fraction may lead tooverestimates of the amount of leucine incorporated intoproteins. To determine the magnitude of this overestimate,we compared incorporation rates of [3H]leucine followinghot- and cold-TCA extractions over time in samples fromDelaware Bay (Roosevelt Inlet) and the Mid Atlantic Bight.The amount of [3Hlleucine incorporated into cold-TCA-insoluble material was in close agreement with that incorpo-rated into hot-TCA-insoluble material (Fig. 1). The averagedifference was <10% of the [3H]leucine recovered in cold-TCA-insoluble material and was within the experimentalerror of the measurements. This difference was consistent

VOL. 54, 1988

on May 16, 2018 by guest

http://aem.asm

.org/D

ownloaded from

1936 CHIN-LEO AND KIRCHMAN

TABLE 1. Incorporation of [3H]leucine into cold- andhot-TCA-insoluble material during time course experiments in

Delaware Bay and the Mid Atlantic Bight

Datea and Incorporation rate (pM h-1) Ratio of cold/time (min) Cold TCA Hot TCA hot incorporation"

Feb 8630 268 230 1.1760 424 371 1.1490 653 456 1.43120 542 655 0.83

Mar 8630 817 779 1.0560 1,742 1,770 0.9890 2,722 2,552 1.07120 3,669 3,666 1.00150 4,440 4,569 0.97

Jun 8760 329 275 1.2090 311 344 0.90120 434 569 0.76150 563 528 1.07

a The first two experiments were conducted with samples from RooseveltInlet. The third experiment was conducted with samples from the MidAtlantic Bight (74°05' W, 37°05' N).bRatio of rates of [3H]leucine incorporation obtained with a cold-TCA

extraction to those obtained with a hot-TCA extraction. Mean values (-standard deviation) are 1.14 ± 0.25 for February 1986, 1.01 ± 0.04 for March1986, and 0.98 ± 0.19 for June 1987.

throughout the experiment (120 min). Similar results wereobserved in experiments performed at various times andlocations (Table 1). The average ratio of [3H]leucine recov-ered after a cold-TCA extraction to that recovered after ahot-TCA extraction was 1.0 ± 0.2. In the waters tested, acold-TCA extraction did not appear to lead to a substantialoverestimate of the level of [3H]leucine incorporation intoproteins.Comparison of single-label and dual-label incubations. The

dual-label method requires the addition of two differentsubstrates in high concentrations. This organic enrichmentcould potentially affect incorporation by natural bacterialassemblages. To test this possibility, we compared rates ofincorporation obtained by the dual-label method with thoseobtained from single-label incubations at various times insamples taken from various locations in Delaware Bay, theChesapeake Bay, and the Mid Atlantic Bight.Leucine incorporation was generally higher (85% of all

comparisons, n = 48) in dual-label than in single-labelincubations, but this difference was small (average ratio of14C to 3H incorporation, 1.1 ± 0.2 [Table 2]). A scatter plotof all dual- and single-label leucine incorporation valuesyielded a linear relation (Fig. 2) with a correlation factor, r,of 0.96 (n = 48). The slope of the regression line was 1.10 +0.07 (standard error), which is not significantly different fromthe expected slope of 1 (P > 0.05; Student's t test).

During the growth experiments designed to calculateconversion factors, however, leucine incorporation rateswere significantly higher in dual-label than in single-labelincubations. This difference was much greater than thatobserved in normal samples and occurred during the initialincreases in bacterial abundance (Fig. 3).Measurements of [3H]thymidine incorporation by the du-

al-and single-label methods covaried along the axis of Dela-ware Bay (Fig. 4) and were in close agreement. The corre-

TABLE 2. Comparison of leucine incorporation in dual-and single-label incubations at various times and locations in

the Chesapeake Bay

Incorporation of

Date Station~' Timeb leucine (pM h-') Dual/singleDual Single rtolabel label

Oct 86 848 1630 856 1,070 0.80848 2100 765 696 1.16848 0120 726 726 1.00848 0500 260 372 0.65848 0900 337 260 1.26848 1300 274 247 0.77848 1700 435 483 1.29

Jun 87 927 0600 244 188 1.30653 0757 12 11 1.13744 0220 128 104 1.24859 1908 100 97 1.03

a Station nomenclature follows the Chesapeake Bay Institute system; digitsrefer to the degree and minutes of the station longitude. These stations are inthe main channel of the bay.

b Times are all Eastern Standard Time.' The mean ratio (± standard deviation) was 1.10 ± 0.20.

lation coefficient for these values was 0.93 (n = 9). Noobvious trend was observed in the differences between thesemeasurements; the average ratio of thymidine incorporationobtained by the dual-label procedure to that obtained by thesingle-label approach was 1.0 ± 0.1.Comparisons of rates of leucine and thymidine incorpora-

tion obtained by the dual- and single-label procedures indi-cate that for the environments tested, the presence of bothleucine and thymidine in the dual-label method does notsignificantly affect their individual incorporation.Conversion factors. Conversion factors for [3H]thymidine

and ['4H]leucine by using the dual-label method and of[3H]leucine by using the single-label method were calculatedfor a sample from the Mid Atlantic Bight and for samplesobtained at various locations and depths in the ChesapeakeBay (Table 3). The average thymidine conversion factor was(2.83 ± 1.19) x 1018 cells mol-1, which is similar to thatcommonly used (2.0 x 1018 cells mol-1) (5) and to that

3.0

Eo

0

1.01-

0

0.00 0.4 0.8 1.2 1.6 2 2.4 2.8

[3H] Leu Incorporation (nmol l-'h-1)

FIG. 2. Comparison of [14C]leucine and [3H]leucine incorpora-tion rates. The dashed line was obtained by linear regression of allvalues (slope = 1.10 + 0.07; r = 0.96; n = 35). The solid line has aslope of 1.

APPL. ENVIRON. MICROBIOL.

on May 16, 2018 by guest

http://aem.asm

.org/D

ownloaded from

BACTERIAL PRODUCTION IN MARINE WATERS 1937

8

7

-6

E85

Q 4U

a 3

2

.-6

A

Isotope Incorporation (pmol 7t1h-l)20 40 60 80 100 120

4 -

7

5 9

4,

3

8

10

12E

14

160 6 11 18 24 38.5 47 53

Time (hr)

FIG. 3. Comparison between increases in bacterial abundanceover time and the ratio of ["4C]leucine to [3H]leucine incorporation.Measurements were performed on a surface sample collected fromthe upper Chesapeake Bay.

measured specifically for the Chesapeake Bay (4.0 x 1018cells mol-'; H. Ducklow, personal communication). Theaverage leucine conversion factors were (1.42 ± 0.69) x 1017and (1.64 + 1.21) x 1017 cells mol-1 for the dual- andsingle-label methods, respectively (Table 3).

Application of the method to aquatic environments. Thedual-label method was applied to samples from the Chesa-peake Bay during October 1986 to estimate rates of bacterialproduction. Leucine and thymidine incorporation rates mea-sured by the dual-label method covaried over depth. Subse-quent measurements at the same station over a diel cycleshowed that leucine and thymidine levels continued tocovary through time and that surface values obtained by thedual- and single-label methods were in close agreement. Insurface samples the ratio of [14C]leucine to [3H]thymidineincorporation was 0.80 (Table 2). Similar trends were ob-served in samples from Delaware Bay and the Mid AtlanticBight. These comparisons are part of a larger data base thatwill be discussed in detail elsewhere (G. Chin-Leo and D. L.Kirchman, manuscript in preparation).We applied our measured conversion factors to the depth

120

100!-

80

= 600

9 40

,WM 20

-10 0 10 30 50 70

Distance Upstream (km)

FIG. 4. Incorporation of [3H]thymidine (TdR) in dual- and sin-gle-label incubations as a function of distance from the head ofDelaware Bay. Experiments were performed on surface samplesduring July 1987. The zero distance represents the mouth of the Bay.

18

20

22

24

FIG. 5. Vertical profile of [3H]thymidine (TdR) and ["4C]leucine(Leu) incorporation obtained by the dual-label method at station 848in Chesapeake Bay (76°25' W, 38°48' N) during October 1986.['4C]leucine incorporation rates should be multiplied by 10.

profile of incorporation rates given in Fig. 5. Data onthymidine and leucine incorporation indicated that bacterialproduction was highest in surface and bottom waters (Table4). The production rate in surface waters was estimated to be13.3 x 107 and 10.8 x 107 cells liter-' h-1 based on[3H]thymidine and [14C]leucine incorporation, respectively.Overall, the difference between production estimates basedon thymidine and leucine incorporation was small (<25%)and was within experimental errors. Since our measuredconversion factor (2.83 x 1018 cells mol-') was very similarto the commonly used factor of 2.0 x 1018 cells mol-' (6),bacterial production estimates from these factors were alsovery similar (Table 4).

DISCUSSION

Our results indicate that in the coastal waters tested,thymidine and leucine incorporation can be measured in asingle incubation by the dual-label approach and that thymi-dine and leucine incorporation give similar estimates ofbacterial production. Incorporation rates determined fromdual-label incubations differed only slightly from those ob-tained from single-label incubations; for leucine, the differ-ences can probably be attributed to the radioactivity inleucyl-tRNA which is retained in the cold-TCA-insolublefraction. The size of this pool did not seem to increase withincubation time (Fig. 1), probably owing to rapid incorpora-tion of radiolabeled leucine into proteins (16). Addition oftwo substrates in relatively high concentrations (25 nM) didnot appear to stimulate bacterial production during shortincubations (<5 h).An anomalous observation was that during growth exper-

iments [14C]leucine incorporation measured by the dual-label technique (cold-TCA extraction) was two to threetimes higher than the single-label [3H]leucine incorporationrate (hot-TCA extraction) (Fig. 2). The ratio of [14C]leucineto [3H]leucine incorporation was high only during the initialincrease in cell numbers (Fig. 3). A possible explanation is

VOL. 54, 1988

on May 16, 2018 by guest

http://aem.asm

.org/D

ownloaded from

1938 CHIN-LEO AND KIRCHMAN

TABLE 3. Comparison of conversion factors calculated from growth experiments in the Mid Atlantic Bight and the Chesapeake Bay

Date and Conversion factor" (cells mol-) for:location (depth) [3H]TdR (dual label) ['4C]Leu (dual label) [3HILeu (single label)

May 87MABb (3.24 ± 1.25) x 1018 (1.20 ± 0.44) x 10'7 (1.89 ± 0.76) x 107CB" (2 m) (5.65 ± 2.62) x 1018 (1.28 ± 1.06) x 1017 (3.10 ± 2.60) x 1017

July 87CBd (2 m) (1.42 ± 0.67) x 1018 (0.34 ± 0.15) x 1017 (0.43 ± 0.19) x 1017CBd (28 m) (1.00 ± 0.23) x 1018 (2.86 ± 1.09) x 1017 (1.14 ± 0.13) x 1017" Average conversion factors (± standard error) were (2.83 ± 1.19) x 1018 for [3H]thymidine ([3HITdR), (1.42 ± 0.69) x 1017 for [14C]Leu, and (1.64 ± 1.21)

x 1017 for [3H]Leu.b Samples were obtained in the Mid Atlantic Bight (MAB) at a depth of 2 m at 75°16.54' W, 37'02.86' N.' Samples were obtained in the Chesapeake Bay (CB) at station 848 (76°25' W, 38'48' N).d Samples were obtained in the Chesapeake Bay at station 858 (76°22' W, 38°58' N).

that during periods of increasing growth rate, assimilation ofleucine into leucyl-tRNA may be faster than its incorpora-tion into proteins, thus leading to an accumulation of radio-activity in the cold-TCA-insoluble fraction.With the thymidine method, accurate estimates of bacte-

rial production depend on using the appropriate conversionfactor. Our growth experiments provide more evidence thatthe average conversion factor is near 2.0 x 1018 cells mol'(6). However, we also observed almost a sixfold variation inthis factor. For many questions, highly accurate estimates ofbacterial production are not needed, nor is it even possible tomeasure conversion factors for each production estimate.For example, the ability to determine the spatial and tempo-ral variation in bacterial production over scales of hours andkilometers does not depend on knowing the absolute rate ofproduction. Estimating conversion factors every hour orkilometer would be impossible. However, it is necessary toensure that the observed changes in thymidine incorporationare due to real changes in bacterial production, not in theconversion factor or some artifact. Because two independentincorporation rates are measured simultaneously, the dual-label approach is more likely to separate real changes fromartifacts than the single-label approach.The dual-label approach is useful as a method for measur-

ing bacterial production only when thymidine and leucineincorporation covary, which occurs frequently (see above;Chin-Leo and Kirchman, in preparation). However, in-stances when thymidine and leucine incorporation failed to

TABLE 4. Estimates of bacterial production calculated fromincorporation rates of [3H]thymidine and ['4C]leucine"

Production (107 cells liter-' h-')Depth (m)

calculated from incorporation of': ["4C]Leu/[3H]TdR [3H]TdR 14 [3HlTdR'(ref. 6) (this study) ["C]Leu

4 9.4 13.3 10.8 0.816 5.2 7.4 7.0 0.95

15 2.5 3.6 3.8 1.1018 5.0 7.1 6.0 0.8524 10.0 14.0 8.0 0.57

aIncorporation rates were measured by the dual-label method with samplesobtained at station 848 in the Chesapeake Bay in October 1986.

b Conversion factors used were 2 x 10l8 cells per mol for [3H]thymidine([VH]TdR) (obtained from reference 6), 2.83 x 1018 cells per mol for[3H]thymidine (this study), and 1.42 x 10i7 cells per mol for ['4C]leucine (thisstudy).

< Ratios were calculated by using the thymidine and leucine conversionfactors obtained in this study. The mean ratio (± standard deviation) was 0.86± 0.19.

covary have also been observed. Specific problems witheither method may explain this lack of covariance. Method-ological problems of the thymidine procedure that may leadto erroneous production estimates include incorporation ofthymidine into protein and RNA (18), extensive catabolismof thymidine in oxic surface waters (12a), and a significantfraction of the growing bacterial assemblage not taking upthymidine (3, 19). A potential source of artifact of the leucineprocedure is turnover of protein, although rates in naturalassemblages published to date have been low (17). Further-more, although a high percentage (ca. 90%) of labeledleucine has been reported to be incorporated into protein(14), more studies are needed to test the generality of thispattern. Finally, isotope dilution by internal and externalpools of leucine and thymidine can alter estimates of bacte-rial production by either technique.

Alternatively, lack of covariance may be explained byperiods of unbalanced growth when rates of macromolecularsynthesis are uncoupled (D. L. Kirchman, S. Y. Newell, andR. E. Hodson, Int. Symp. Microb. Ecol. 1985, in press).This phenomenon has been observed in pure cultures andstudied extensively (13); it generally occurs when bacteriashift from one growth rate to another. For example, duringan increase in growth rate, the rate of RNA synthesis firstincreases; this is followed by increases in protein and DNAsynthesis, leading eventually to increases in cell division.Since the thymidine and leucine methods estimate DNA andprotein synthesis, respectively, lack of covariance mayreflect unbalanced growth.

In conclusion, the dual-label procedure greatly reducesthe time and materials needed to routinely perform simulta-neous measurements of thymidine and leucine incorporationwhile minimizing errors associated with repeated subsamp-ling and sample manipulation. Coincident measurements ofleucine incorporation provide an independent check of thethymidine method and thus aid in circumventing the uncer-tainties associated with unknown variations in the conver-sion factor. This qualitative approach may be useful whenstudying variations in bacterial production over time andover small distances where conversion factors may vary.Furthermore, lack of covariance between thymidine andleucine incorporation may indicate periods of unbalancedgrowth when rates of macromolecule syntheses are uncou-pled (10, 11). These events may reflect periods of shiftinggrowth conditions, and their detection can be useful in thestudy of the metabolism of bacteria and its regulation byenvironmental conditions (14; Chin-Leo and Kirchman, inpreparation).

APPL. ENVIRON. MICROBIOL.

on May 16, 2018 by guest

http://aem.asm

.org/D

ownloaded from

BACTERIAL PRODUCTION IN MARINE WATERS 1939

ACKNOWLEDGMENTSWe thank R. Biggs for his support and encouragement. We are

grateful to R. Rivkin for permitting G. Chin-Leo to participate on theCorsair 12 cruise aboard RV Cape Hatteras.

This research was supported in part by National Science Foun-dation grants OCE 64554165 (awarded to M. Tyler, R. Biggs, and L.Harding), OCE 8314607, OCE 8520278 (awarded to R. Rivkin), andOCE 8614170 and National Oceanographic and Atmospheric Ad-ministration sea grant NA 86AA-D-SG040 (awarded to D. L. Kirch-man).

LITERATURE CITED1. Bell, T. B., G. M. Ahlgren, and I. Ahlgren. 1983. Estimating

bacterioplankton production by measuring [3H]thymidine incor-poration in a eutrophic Swedish lake. Appl. Environ. Microbiol.45:1709-1721.

2. Bevington, P. R. 1969. Data reduction and error analysis for thephysical sciences. McGraw-Hill Book Co., New York.

3. Douglas, J. D., J. A. Novitsky, and R. 0. Fournier. 1987.Microautoradiography-based enumeration of bacteria with esti-mates of thymidine-specific growth and production rates. Mar.Ecol. Prog. Ser. 36:91-99.

4. Ducklow, H. W., and S. Hill. 1985. The growth of heterotrophicbacteria in the surface waters of warm core rings. Limnol.Oceanogr. 30:239-249.

5. Fuhrman, J. A., and F. Azam. 1980. Bacterioplankton second-ary production estimates for coastal waters of British Columbia,Antarctica, and California. Appl. Environ. Microbiol. 39:1085-1095.

6. Fuhrman, J. A., and F. Azam. 1982. Thymidine incorporation asa measure of heterotrophic bacterioplankton production inmarine surface waters: evaluation and field results. Mar. Biol.(Berlin) 66:109-120.

7. Fuhrman, J. A., R. W. Eppley, A. Hagstrom, and F. Azam. 1985.Diel variation in bacterioplankton, and related parameters in theSouthern California Bight. Mar. Ecol. Prog. Ser. 27:9-20.

8. Grivell, A. R., and J. F. Jackson. 1968. Thymidine kinase:evidence for its absence from Neurospera crassa and someother microorganisms, and the relevance of this to the specificlabelling of deoxyribonucleic acid. J. Gen. Microbiol. 54:307-317.

9. Hagstrom, A. 1984. Aquatic bacteria: measurements and signif-icance of growth, p. 495-501. In M. J. Klug and C. A. Reddy(ed.), Current perspectives in microbial ecology. AmericanSociety for Microbiology, Washington, D.C.

10. Hanson, R. B., and H. K. Lowery. 1983. Nucleic acid synthesisin oceanic microplankton from the Drake Passage, Antarctica:evaluation of steady-state growth. Mar. Biol. (Berlin) 73:79-89.

11. Hanson, R. B., L. R. Pomeroy, and R. E. Murray. 1986.Microbial growth rates in a cold-core Gulf Stream eddy of thenorthwestern Sargasso Sea. Deep-Sea Res. 33:427-446.

12. Hobbie, J. E., R. J. Daley, and S. Jasper. 1977. Use ofNuclepore filters for counting bacteria by fluorescence micros-copy. Appl. Environ. Microbiol. 33:1225-1228.

12a.Hollibaugh, J. T. 1988. Limitations of the [3H]thymidine methodfor estimating bacterial productivity due to thymidine metabo-lism. Mar. Ecol. Prog. Ser. 43:19-30.

13. Ingraham, J. L., 0. Maaloe, and F. C. Neidhardt. 1983. Growthof the bacterial cell. Sinauer Associates, Sunderland, Mass.

14. Kirchman, D. L., E. K'nees, and R. Hodson. 1985. Leucineincorporation and its potential as a measure of protein synthesisby bacteria in natural aquatic systems. Appl. Environ. Micro-biol. 49:599-607.

15. Kirchman, D. L., H. W. Ducklow, and R. Mitchell. 1982.Estimates of bacterial growth from changes in uptake rates andbiomass. Appl. Environ. Microbiol. 44:1296-1307.

16. Kirchman, D. L., and R. E. Hodson. 1986. Metabolic regulationof amino acid uptake in marine waters. Limnol. Oceanogr. 31:339-350.

17. Kirchman, D. L., S. Y. Newell, and R. E. Hodson. 1986.Incorporation versus biosynthesis of leucine: implications formeasuring rates of protein synthesis and biomass production bybacteria in marine systems. Mar. Ecol. Prog. Ser. 32:47-59.

18. McDonough, R. J., R. W. Sanders, K. G. Porter, and D. L.Kirchman. 1986. Depth distribution of bacterial production in astratified lake with an anoxic hypolimnion. Appl. Environ.Microbiol. 52:992-1000.

19. Ramsay, A. J. 1974. The use of autoradiography to determinethe proportion of bacteria metabolizing in an aquatic environ-ment. J. Gen. Microbiol. 80:363-373.

20. Riemann, B. 1985. Potential importance of fish predation andzooplankton grazing on natural populations of freshwater bac-teria. Appl. Environ. Microbiol. 50:187-193.

21. Riemann, B., P. K. Bjornsen, S. Newell, and R. Fallon. 1987.Calculation of cell production of coastal marine bacteria basedon measured incorporation of [3H]thymidine. Limnol. Ocea-nogr. 32:471-476.

22. Tremaine, S. C., and A. L. Mills. 1987. Tests of the criticalassumptions of the dilution method for estimating bacterivoryby microeucaryotes. Appl. Environ. Microbiol. 53:2914-2921.

VOL. 54, 1988

on May 16, 2018 by guest

http://aem.asm

.org/D

ownloaded from