assimilatory sulfur metabolism microorganisms ... · protein synthesis in unenriched...

Vol. 43, No. 1APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Jan. 1982, p. 160-1680099-2240/82/010160-09$02.00/0

Assimilatory Sulfur Metabolism in Marine Microorganisms:Considerations for the Application of Sulfate Incorporationinto Protein as a Measurement of Natural Population Protein

SynthesistRUSSELL L. CUHEL,** CRAIG D. TAYLOR, AND HOLGER W. JANNASCH

Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543

Received 12 February 1981/Accepted 28 August 1981

The sulfur content of residue protein was determined for pure cultures ofNitrosococcus oceanus, Desulfovibrio salexigens, 4 mixed populations of fer-mentative bacteria, 22 samples from mixed natural population enrichments, and11 nutritionally and morphologically distinct isolates from enrichments of Sargas-so Sea water. The average 1.09 + 0.14 % (by weight) S in protein for 13 purecultures agrees with the 1.1% calculated from average protein composition. Anoperational value encompassing all mixed population and pure culture measure-ments has a coefficient of variation of only 15.1% (n = 41). Short-term [35S]sulfateincorporation kinetics by Pseudomonas halodurans and Alteromonas luteo-violaceus demonstrated a rapid appearance of 35S in the residue protein fractionwhich was well modelled by a simple exponential uptake equation. This indicatesthat little error in protein synthesis determination results from isotope dilution byendogenous pools of sulfur-containing compounds. Methionine effectively com-peted with sulfate for protein synthesis in P. halodurans at high concentrations (10pLM), but had much less influence at 1 ,uM. Cystine competed less effectively withsulfate, and glutathione did not detectably reduce sulfate-S incorporation intoprotein. [35S]sulfate incorporation was compared with [14C]glucose assimilation ina eutrophic brackish-water environment. Both tracers yielded similar results forthe first 8 h of incubation, but a secondary growth phase was observed only with35S. Redistribution of 14C from low-molecular-weight materials into residueprotein indicated additional protein synthesis. [35S]sulfate incorporation intoresidue protein by marine bacteria can be used to quantitatively measure bacterialprotein synthesis in unenriched mixed populations of marine bacteria.

The direct measurement of true bacterialgrowth rate or metabolic processes has beenseriously hampered by the extreme complexityof aquatic environments in terms of both micro-bial community composition and the physicaland chemical characteristics of the habitats. Theresult is that few quantitative short-term assess-ments of total bacterial growth exist.

Little attention has been paid to the utilizationof inorganic nutrients by bacteria, although theyrequire N, P, and S as do phytoplankton. Inor-ganic nutrient assimilation by marine bacteriamay provide a much needed means of quantify-ing bacterial biomass production. For example,sulfate incorporation into bacterial protein hasbeen shown to be an accurate measure of de

t Contribution no. 4814 from the Woods Hole Oceano-graphic Institution.

t Present address: National Oceanographic and Atmospher-ic Administration, Ocean Chemistry and Biology Laboratory,AOML, Miami, FL 33149.

novo protein synthesis in pure cultures of ma-rine bacteria (7b, 8). In addition to its utility as alaboratory tool for pure culture studies, severalfeatures of the sulfate incorporation measure-ment recommend its use for measurement ofnatural bacterial population protein synthesis.The high sulfate concentration of seawater

can be readily measured and ensures that sulfuris never limiting microbial growth in marineenvironments. The concentration of dissolvedsulfur-containing amino acids in seawater is inthe undetectable-to-low nanomolar range (6, 11,25, 29, 36), suggesting that sulfate is the onlysignificant source of sulfur for bacterial proteinsynthesis, although measurements of S-aminoacid turnover times need to be made to estimatetheir true contribution to bacterial sulfur uptake.Although reductive assimilation of sulfate intoprotein is restricted to microorganisms (i.e.,bacteria, fungi, and green plants), it appears tobe universal among them, thereby functionally

160

on May 2, 2020 by guest

http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

APPLICATION OF SULFUR INCORPORATION BY BACTERIA 161

differentiating the entire microbial communityfrom protozoa and higher animals (30). The poolsize of sulfur-containing protein precursors isalso extremely small in bacteria (3, 7a, 7b, 34,35), minimizing complications arising from labeldilution by large endogenous pools (20).The use of sulfate incorporation into bacterial

protein as an assessment of natural populationprotein synthesis requires that the sulfur contentof protein be relatively constant among bacteria.It is also necessary to determine whether sulfateincorporation measurements are significantly in-fluenced by equilibration of endogenous pools(20), since sufficiently sensitive techniques formeasurement of the intracellular specific activi-ty of sulfur-containing amino acids in naturalpopulations have not been developed. Finally,an understanding of competition for sulfate me-tabolism by reduced sulfur compounds is usefulwhen measurements in natural waters are madenear highly productive or anaerobic regionswhere reduced sulfur compounds exist.

This paper presents the results of a survey ofthe sulfur content of protein from natural popu-lation enrichments and pure cultures of marinebacteria representing diverse nutritional modesof metabolism. Short-term incorporation kinet-ics and competition with reduced sulfur com-pounds were investigated, and the method wasapplied to a highly eutrophic brackish-waterenvironment as an initial field trial.

MATERIAULS AND METHODSNatural population enrichments. Enrichments of

Sargasso Sea water were made on RIV Oceanuscruises 69 and 75 in the vicinity of 37°28.7' N, 64003.4'W. Samples taken from below the euphotic zone (500m) with EtOH-washed Niskin bottles were prefilteredthrough 28-,um-mesh Nitex net to remove particles,and 100-ml aliquots were dispensed into sterile flasks.Inorganic nutrients (12) and organic carbon com-pounds were added as concentrated stock solutionswhich had been sterilized by autoclaving or filtrationthrough 0.2-,um filters as appropriate, and the flaskswere shaken at 22 ± 2°C at 250 rpm on a Gyrotoryshaker (New Brunswick Scientific Co., Edison, N.J.).Pure cultures were isolated by spreading dilutions ofthe enrichments on seawater medium containing inor-ganic nutrients and the growth substrate, solidified by1.5% agar (Difco Laboratories, Detroit, Mich.). Colo-nies were picked and restreaked several times, and theisolates were maintained on slants of the same medi-um.

Seventy of the isolates obtained on RIV Oceanuscruise 75 were screened for nutritional capability bystreaking onto a series of seawater plates, each con-taining one of 25 individual carbon compounds. Liquidcultures were used for nitrate and ethanol utilization.Based on colony morphology, microscopic observa-tion, and the ability to grow on the various carbon andenergy sources, 11 nutritionally and morphologicallydistinct strains were obtained.

Determination of the sulfur content of protein. (i)Natural populations. The enrichment series from RIVOceanus cruise 69 was labeled with Na235SO42- (10pCi ml-'; 0.8-dpm pmol-' final specific activity) formeasurement of sulfate incorporation into protein.Paired samples (20 to 25 ml) were brought to a finalconcentration of 10% trichloroacetic acid at the firstsign of turbidity and again when the culture was visiblydense (optical density of 420 nm, about 0.6). Analiquot was streaked onto solid medium of the samecomposition to check colony morphology and domi-nance at each sampling. The fixed samples werecentrifuged (5,000 x g, 20 min), and the pellets wererinsed twice with 10% trichloroacetic acid. One sam-ple from each pair was fractionated for the determina-tion of radioactivity in the protein residue (7a), and theother was dissolved in 0.1 N NaOH for protein assay.Good growth was obtained on 12 different substrates.

Fermentative bacteria were enriched from an anoxicestuarine sediment sample (S. Henrichs, Ph.D. thesis,Woods Hole Oceanographic Institution, Woods Hole,Mass., 1980) in a similar manner except that the mediawere sparged with argon for 10 min and incubationswere carried out in sealed tubes. Glucose, galactose,fructose, and mannitol were used as substrates at 20mM final concentrations.

(ii) Pure cultures. Nitrosococcus oceanus was kindlyprovided by S. Watson, Woods Hole OceanographicInstitution. It was grown in RLC-water medium (7)containing 20 mM NH4Cl, 1 mM Na235SO4 (2 dpmpmol-'), and 0.1% phenol red. The pH was adjustedas necessary with 0.1 M K2CO3. The late-exponential-phase culture was harvested by centrifugation, thepellet was rinsed twice with RLC-water, and duplicatesamples were taken for fractionation and protein as-say.

Desulfovibrio salexigens (obtained from J. R. Post-gate, University of Sussex, Sussex, U.K.) was grownanaerobically in RLC-water medium containing 10mM lactic acid and 25 mM Na235SO4 (0.5 dpmpmol-1). A total of 15 samples were taken for fraction-ation and protein assay during exponential- and early-stationary-phase growth in two separate experiments.

Cruise isolates were grown in RLC-water mediumcontaining 1 mM Na2SO4 (2 dpm pmol-') and 10 mMsodium glutamate, sodium succinate, or glucose as thesole carbon and energy sources. The cultures wereincubated aerobically with shaking and were sampledin the late exponential phase of growth. Samples for"5S distribution were filtered onto Whatman GF/Ffilters and rinsed three times with 2 ml of 0.5 M NaCl.The punch funnel described previously (7) was usedduring ffitration to reduce isotope background.

(I) Other methods. Protein, dissolved in 0.1 NNaOH, was measured on the 10%o trichloroacetic acid-insoluble material precipitated directly from the cellsuspension and on the protein residue from the frac-tionation procedure by the method of Bradford (4),using bovine serum albumin as a standard. Directcounts were made by epifluorescence microscopy (9).The sulfate concentration of natural seawater wascalculated from the chlorinity determined on aBuchler-Cotlove chloridometer (Buchler Instruments,Fort Lee, N.J.), using a sulfate/chloride ratio of 0.14(28).

Calculation of the sulfur content of protein. Thesubcellular fractionation procedure used in this study

VOL. 43, 1982


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

162 CUHEL, TAYLOR, AND JANNASCH

TABLE 1. Total protein, total sulfur, and weight percent sulfur in protein of marine bacteriaaPer 108 cells S (wt % in protein)

Expt Total protein (pg) Total sulfur (ng) Operational' True

Mixed-population enrichments NDc ND 0.94 ± 0.13 (22) NDFermentative enrichments ND ND 0.92 ± 0.15 (4) NDEnrichment isolates 16.7 ± 10.3 (8) 274 ± 142 (8) 0.91 ± 0.15 (12) 1.10 ± 0.14 (9)N. oceanus 142.7 2,122 0.78 1.31D. salexigens 9.6 125 1.10 ± 0.15 (15) 1.00 ± 0.06 (6)P. halodurans 23.8 ± 11.6 (162) 292 ± 129 (110) 0.86 ± 0.12 (80) 1.07 ± 0.17 (60)A. luteo-violaceus 12.9 ± 4.9 (79) 106 ± 32 (80) 0.72 ± 0.11 (45) 0.92 ± 0.15 (10)

Overall mean 0.93 ± 0.14 (42) 1.09 ± 0.14 (13)a Experimental procedures are given in the text. Data for P. halodurans and A. luteo-violaceus are taken from

Cuhel et al., (8). The mean ± 1 SD is given for the number of determinations in parentheses. Overall means arefor the indicated number of individual organisms and populations.

b As defined in the text.c ND, Not determined.

permits the determination of the amount of 3IS incor-porated into residue protein in natural populationsamples. However, about 10 to 15% of both the totalcellular protein (by assay) and macromolecular sulfurare solubilized in other fractions, principally the alco-hol-soluble, ether-insoluble component ("alcohol-sol-uble protein") (7a, 27). Thus, it is necessary to definean operational relationship between the residue pro-tein-S and total protein synthesis as well as to deter-mine the true sulfur content of protein. The operation-al weight percent S used in this work is (residueprotein-S/total protein) x 100. When sufficient residueprotein was available for assay in pure culture studies,the true weight percent sulfur in protein could becalculated as (residue protein-S/residue protein) x100.

Radiochemicals. [U-14C]glucose (255 mCi mmol-1)and Na235SO4 (carrier-free) were obtained from Amer-sham Corp. (Chicago, Ill.).Other chemicals. Reduced glutathione was obtained

from Sigma Chemical Co. (St. Louis, Mo.). Aquasolwas purchased from New England Nuclear Corp.(Boston, Mass.). All other chemicals were reagentgrade.

RESULTSSurvey of the sulfur content of protein in ma-

rine bacteria. Enrichment of Sargasso Sea waterwith a variety of individual carbon and energysources was performed to ascertain the protein-sulfur/total protein relationship for organismsresponding to different growth substrates.Streaking of the cultures on solid medium gaverise to numerous colony morphologies in mostflasks, indicating that the residue protein-S/totalprotein ratio was not dominated by protein fromsingle species.To further verify the applicability of the pro-

tein-S analysis to natural populations, isolatesfrom a later enrichment series (RIV Oceanuscruise 75) were grown in pure culture, and thecell number, sulfur, and protein relationshipswere determined under completely defined con-ditions. These parameters were also measured

for an anaerobic sulfate-reducing bacterium, D.salexigens, a marine nitrifying bacterium, N.oceanus, and four mixed populations offerment-ative bacteria enriched from a mud sample fromthe Pettaquamscutt River estuary (Kingston,R.I.) during cooperative experiments with otherlaboratories at Woods Hole. The results of theseanalyses are summarized in Table 1. The greatvariability in cellular protein content of theisolates was accompanied by a similar range ofprotein-S, resulting in a relatively constant sul-fur content of protein (coefficient of variation,13.1%).A key difference between bacterial and algal

sulfate assimilation is the synthesis of sulfolipidby plants (cf. Busby and Benson [5]); hence,knowledge of the distribution of sulfur in themajor biochemical fractions is of interest. Table2 shows the averages for cruise isolates, Altero-monas luteo-violaceus, N. oceanus, and Pseu-domonas halodurans. One cruise isolate, no. 50,stood out from the others by virtue of its highproportion of inorganic sulfate and low-molecu-lar-weight (LMW) organic sulfur. Only in N.oceanus, however, was there a significantamount of lipid-S. The four populations of fer-mentative bacteria displayed an unusually highproportion of protein-S (>90% of the total 35S).The weight percent S in protein and sulfur

distribution for each isolate or enrichment aretabulated in a thesis (R. L. Cuhel, Ph.D. thesis,Woods Hole Oceanographic Institution, WoodsHole, Mass., 1981).

Effects of low concentrations of compoundscontaining reduced sulfur on sulfate incorpo-ration by P. halodurans. A consideration in theuse of sulfate incorporation into protein as ameasurement of marine microbial protein syn-thesis is the possible effect of other sulfur-containing compounds (e.g., cysteine, methio-nine, glutathione, and thiosulfate). Sulfate mustbe reduced before entering into the protein bio-

APPL. ENVIRON. MICROBIOL.


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


TABLE 2. Distribution of 35S in biochemical fractions of marine bacteria in pure culturea% of total radioactivity

Organism SO 2- LMW Alcohol Lipid Hot TCAb Residue504 organic soluble Lpi soluble protein

N. oceanus 2.0 19.4 8.3 4.3 13.4 52.6P. halodurans 0.9 13.3 4.1 1.7 8.6 71.4A. luteo-violaceus 3.2 17.9 3.0 0.9 4.9 70.1Cruise isolates 2.7 18.5 11.7 1.7 7.4 59.6Isolate 50 5.6 44.8 6.7 0.8 5.9 36.2

a The mean distribution for cruise isolates is shown. Data for P. halodurans and A. Iuteo-violaceus were takenfrom previous work (8).

b TCA, Trichloroacetic acid.

synthetic pathway, and preferential utilization ofnaturally occurring sulfur compounds wouldlower apparent sulfur incorporation measuredwith sulfate. It is therefore desirable to know atwhat concentration reduced sulfur compoundsaffect the incorporation of sulfate sulfur intoprotein.

P. halodurans was grown in complete mediumcontaining 1 mM sulfate supplemented with lowconcentrations of cystine (1 and 4.5 ,uM, equalto 2 and 9 ,uM S, respectively), methionine (1and 10 ,uM), glutathione (1 and 10 ,uM), thiosul-fate (1 and 10 FLM), and a control without addedsulfur. The supplements had no effect on thegrowth rate. During late-exponential-phasegrowth (ca. 5 x 107 cells ml-'), methionine andcystine clearly affected the amount of sulfate-Sappearing in LMW organic pools and residueprotein-S (Table 3). Methionine caused a redis-tribution of sulfur from residue protein-S to theLMW organic-S fraction, and the effect wasproportional to the concentration of added sul-fur. In contrast, cystine suppressed incorpo-ration of sulfate into the LMW organic-S pool,thus increasing the proportion of sulfate-S inresidue protein. A similar pattern of sulfate-sulfur distribution was observed in the earlystationary phase (ca. 2 x 108 cells ml-'), butonly at the higher level of added sulfur. Thiosul-fate and glutathione were without appreciable

effect at these concentrations.The effect of cystine and methionine observed

in the distribution of sulfur was amplified byanalysis of the sulfate-sulfur content of the totaland residue protein. Table 3 emphasizes theeffect of methionine on sulfate incorporation; atthe 10 ,uM concentration over two-thirds of theresidue protein sulfur was derived from methio-nine early in growth, and the effect remained,although damped, at the higher cell density. Aproportional reduction in the residue protein-S/residue protein ratio was observed. The lowerconcentration of methionine exerted a barelydetectable effect. The effect of cystine at 1 ,uMwas more pronounced than the effect of methio-nine at the same concentration.

Short-term sulfate incorporation kinetics. Ap-plication of sulfate incorporation into protein asa method for measuring bacterial protein synthe-sis requires that there be little lag in the incorpo-ration of sulfate into protein due to the equilibra-tion of radiosulfate with endogenous pools (cf.Karl [20]). Data illustrating this to be the case forP. halodurans and A. luteo-violaceus are shownin Table 4. The experiments are based upon thefact that cells in a balanced state ofgrowth are ofconstant cellular composition in time and arecharacterized by constant specific rates of poly-mer synthesis. Expression of these properties interms of isotope incorporation into polymers,

TABLE 3. Effects of methionine and cystine on the incorporation of sulfate into LMW organic compoundsand residue protein by P. haloduransa

% of total radioactivitySupplement Level (>m) . . Protein = S (wt %)

LMW organic Residue protein

None 11.9 78.5 0.75Methionine 1 16.1 72.8 0.71

10 35.9 57.1 0.23Cystine 1 9.6 78.8 0.65

4.5 8.4 82.4 0.59

a Complete medium containing 10 mM sodium glutamate and 1 mM Na2 5SO4 (2 dpm pmolP) was inoculatedwith late-exponential-phase cells to a final density of 3 x 104 cells ml-'. Aliquots were transferred to sterileflasks and supplemented with filter-sterilized sulfur sources at the indicated concentration. During the mid- tolate-exponential phase and again in the early stationary phase of growth samples were taken for total protein andbiochemical fractionation. Data are expressed as radioactivity derived from sulfate.

VOL. 43, 1982


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


TABLE 4. Proportion of total 3"S in residue proteinin short-term incubationsa

Time % of total 35S(min) P. halodurans A. luteo-violaceus

5 67.2 50.715 74.3 63.130 75.8 69.8120 75.8 71.8480 63.7 73.1a Experimental details are given in the legend to Fig.

1.

however, will be manifest only when the LMWprecursor pools are in a state of isotope equilib-rium. In P. halodurans over 67% of the totalnewly incorporated S was found in the proteinresidue in 5 min. After 15 min of incubation theproportion of sulfur in the residue protein,74.3%, was essentially constant in time, indicat-ing isotopic equilibration of the precursor poolswithin approximately 0.25 generation. The de-creased proportion of 35S-labeled protein at 480min reflects a condition of imbalanced growth ofP. halodurans and resulted from the abruptdecrease in the rate of protein synthesis afterapproximately 2.5 h of incubation (Fig. 1).A similar pattern of incorporation was ob-

served for A. Iuteo-violaceus, where in 30 min(approximately 0.25 generation) the proportionof 35S in residue protein was within 6% of themaximum observed.Under conditions of balanced growth and rap-

id precursor equilibrium the accumulation ofisotope into specific cellular substituents (I) maybe described by the equation (27):

- QO (eklf - 1)k 1n2

Qo is the initial rate of isotope incorporation andk is the reciprocal doubling time. A semilog plotof isotope accumulation versus time (t) is con-vex, and asymptotically approaches a straightline of slope k ln2. The degree of curvature isdirectly related to k and can be used to deter-mine the growth rate of exponentially growingpure cultures (32, 33). Examined in this manner,the isotope uptake data in Fig. 1 are well de-scribed by equation 1 throughout the exponen-tial period of growth. For comparison, totalprotein and direct cell counts are also shown.For P. halodurans growth rate constants com-

puted from total uptake of 35S and incorporationinto residue protein was 0.97 ± 0.01 and 1.01 +0.02 h-1, respectively, over the time period of 0to 2.5 h. The values for k obtained from cellcounts and total protein were 1.03 and 1.29 h-',respectively. Although for A. luteo-violaceus

the growth rate determined for direct counts(0.48 h-1) was less than that for total protein(0.57 h-1), there was no deviation from expo-nentiality for nearly 8 h of growth. In this casethe best-fit k values for all of the isotope incor-poration data from 5 min to 8 h were determinedto be 0.59 ± 0.01 h-1 for total 35S uptake and0.63 ± 0.01 h-1 for 35S incorporation into resi-due protein.

Sulfate incorporation in a eutrophic marinesystem. The first field trial of this method wasconducted in a highly eutrophic natural seawatersystem, a holding tank for the estuarine killifish,Fundulus heteroclitus. This environment offeredtwo advantages for initial attempts to measurenatural population protein synthesis rates. First,the environment of the Fundulus is estuarine;hence, the holding tank was kept at half-strengthseawater. The reduced sulfate concentration (14

5000 -

2000 -

1000 -

500 -

200 -

100 -

50 -

20 -

10 -

5 -

2 -

05 -

0.2 -

0 2 3 4 5 6Incubation Time(Hours)

7 8

FIG. 1. Direct counts, total protein, and 35So42-uptake and incorporation kinetics for exponentiallygrowing cultures of P. halodurans and A. luteo-viola-ceus. Complete medium containing 10 mM glutamateand 1 mM sulfate was inoculated with the appropriateorganism at about 2 x 104 cells ml-'. When the celldensity was near 107 cells ml-', Na235SO42- wasadded to a final activity of 4 x 106 dpm ml-' (P.halodurans) or 3 x 106 dpm ml-1 (A. luteo-violaceus),and aliquots were sampled for radioisotope uptake atshort intervals.

0_0-u-- Bulk Protein

0/ 0 Vv: - 0 -o-

v Direct Counts



http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


sulfate doubled in both fractions. LMW organic-S accounted for 20% of the total 35S.The total uptake of glucose did not give any

indication of the second growth phase between 8TAL and 12 h of incubation (Fig. 3). Glucose assimila-

tion increased during the first 3 h to a plateauwhich remained unchanged for the remainder ofthe experiment. Production of CO2 continued toincrease rapidly for 1 h after the cessation ofincorporation of glucose into cell material (data

PROTEIN not shown) and then continued at a slow rate forthe duration of the experiment to achieve a finalmaximum utilization (incorporation plus respira-

A tion) of 63.5% of the total glucose.NIC Fractionation of the 14C-labeled bacteria into

major biochemical components provided indica-tions of the secondary growth phase, eventhough total incorporation had ceased. Figure 3

8 12 shows a rapid increase in all fractions during theIncubation Time (Hours)

FIG. 2. Time course of sulfate assimilation by nat-ural bacterial populations in a brackish-water fishtank. A 2-liter sample from a well-aerated half-strengthseawater (chlorinity, 9.97%o) holding tank for Funduluswas rapidly ifitered through 28-,um Nitex net to re-move coarse particles. One liter was poured into alarge syringe made from a chromatography column (6by 38 cm; Bellco Glass Co., Vineland, N.J.) fitted witha polycarbonate end cap and piston with rubber 0-rings. A 10-mCi amount of carrier-free Na235SO42-(final specific activity, 2 dpm pmol-') was added andmixed by inversion of the syringe. Subsamples wereremoved through a Luer-lock fitting in the end capafter flushing about 5 ml of the sample through theport. Duplicate 50-ml samples were filtered throughWhatman GF/F filters. Error bars are shown whenlarger than the symbol. The initial cell density was 4.4 x106 ml-1, and the sample and incubation temperaturewas 22°C.

mM) was almost 100 times the concentration atwhich sulfate limitation occurs in marine algae(24) or bacteria (7a) but gave a twofold-greatersensitivity to the sulfate incorporation measure-ment. Second, the use of high-food-value fishfeed combined with the excretion of organic andinorganic compounds by the fish resulted in anenvironment highly conducive to healthy bacte-rial growth, verified by the observation of arelatively high cell density (4.4 x 106 cells ml-')of large, curved, rod-shaped bacteria about 2 by5 ,um in dimensions. Incoming seawater flushedthe tank about three times per day, providing asemicontinuous culture regime.The time course of sulfate uptake by the

bacterial population in the fish tank was charac-terized by rapid incorporation into protein andLMW organic-S components during the firsthour followed by an extended plateau (Fig. 2).At 8 to 12 h of incubation, incorporation of

first 3 h, with protein as the major end product ofglucose assimilation. During the stationary peri-od (as determined by whole-cell 14C), movement

30-

25-

a- 20-

~0

0Il

10-

5-

01 234 8 12Incubation Time (Hours)

FIG. 3. Total uptake of [U-14C]glucose and its dis-tribution in major biochemical fractions of bacteriafrom a brackish-water fish tank. Sampling and incuba-tion procedures are described in the legend to Fig. 2.[U-_4C]glucose (250 mCi mmol-1) was added at a finalactivity of 3.7 ,uCi liter-', and 10-ml samples werefiltered for total uptake (0.2-1Lm membrane filters) andfractionation into biochemical components (GF/F fil-ters). Symbols: 0, sum of the activity in each fraction;0, whole-cell unprocessed controls.

3500 -

3000-

Es 2500-cnC/)c. 2000-

=3 1500-

= 1000-

500-

TOI

-----

I L.M.W ORGASA._ -AA

I I I I

01 234

TOTAL c0 °

0

I0

HOT ACID SOLUBLE

// L.M.W - A

o.1 LIPID

VOL. 43, 1982


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


of label from the labile LMW pool into residueprotein (and to a lesser extent hot trichloroaceticacid-soluble material) was evident. The 14C con-tent of lipid, a more structurally related fraction,remained constant during this time.

DISCUSSIONThe results of the survey of the residue pro-

tein-S/total protein ratio for mixed natural popu-lations in enrichment culture and isolates fromsimilar enrichments in pure culture corroboratethe validity of measuring marine bacterial pro-tein synthesis by using sulfur incorporation intoprotein. A wide variety of microorganisms werestudied, including a chemolithotroph (N. oce-anus), a sulfate-reducing bacterium (D. salexi-gens), mixed populations of fermentative andheterotrophic bacteria, and pure cultures of het-erotrophic bacteria with varying degrees of nu-tritional versatility. Few organisms among themdeviated seriously from the mean ratio, and thecoefficient of variation of the residue protein-S/total protein ratio for the cruise isolates, 16.8%,is much less than that for either the protein(61.8%) or the total sulfur (51.7%) per cell.The agreement of the true weight percent

sulfur in protein (i.e., [residue protein-S/residueprotein] x 100) determined for 13 different bac-teria, 1.09%, with the 1.1% calculated from theJukes et al. (19) average protein, is reassuring.The variation in individual protein compositionsummarized by Holmquist (17) is indeed aver-aged out in total protein. In view of the consist-ency of the residue protein-S/total protein ratioamong bacteria and in an individual organismunder a variety of nutritional regimes, it isproposed that the measurement of sulfate incor-poration into residue protein is a quantitativeassay for marine bacterial protein synthesis.The small absolute size of the LMW organic

sulfur pool and the negligible contribution ofinorganic sulfate to the total cellular sulfur fortwo marine bacteria (7a) permit extremely rapidincorporation of 35s042- into residue protein(Table 4). Much of the sulfur amino acid require-ment for protein synthesis can be satisfied bydirect reduction of sulfate rather than utilizationof preexisting precursors. Studies of the molecu-lar organization of the cysteine synthetase sys-tem (22) support this concept, finding that asingle, multimeric complex in Salmonella typhi-murium performed both O-acetylation of serineand sulfurylation of the product O-acetylserinewith S2- to yield cysteine. Likewise, the reduc-tion of adenosine 3'-phosphosulfate to the levelof sulfide occurs in a single six-electron stepwhile bound to an LMW protein carrier (1, 31);the authors suggest that the entire sulfate reduc-tion pathway occurs while enzyme bound. Thefast reaction kinetics of such systems due to

increased substrate concentrations resultingfrom proximity to reaction centers while enzymebound allows the maintenance of small precur-sor pools. Further reactions of cysteine duringmethionine biosynthesis may be similarly coor-dinated, since the intermediates homocysteineand cystathionine are also found at very lowconcentration (15). The slowly equilibratingLMW component may be primarily glutathione,which is well known to be a sulfur reserve inbacteria (2, 10, 13). Glutathione may serve as anemergency sulfur source and oxidation-reduc-tion buffer without substantial involvement as asource of cysteine for amino acid biosynthesis,since it derepresses components of the cysteinebiosynthetic pathway effectively (21).The measurement of sulfate incorporation into

the protein residue therefore represents de novoprotein synthesis within a small degree of errorin growing cells containing a normal comple-ment of sulfur in the LMW pool. This is espe-cially valuable because it minimizes problemsarising from dilution of label by large endoge-nous pools of protein precursors (20) and per-mits the direct interpretation of sulfate incorpo-ration studies in mixed natural populations.The specificity of the incorporation of sulfur

into residue protein for protein synthesis isemphasized by the good fit of incorporation datato the model derived from elementary consider-ations of isotope uptake kinetics. For both P.halodurans and A. luteo-violaceus, determinedvalues of k from total sulfur uptake and incorpo-ration into residue protein agreed within anexperimental error of ±5%. These values com-pared within ±25% of values obtained fromindependent cell count or total protein measure-ments, which in themselves agreed within asimilar magnitude of error. Thus, within thelimits discussed earlier for use of the model,studies of the dynamics of isotope incorporationcan, in the absence of any further ancillaryinformation, provide an estimate of the growthrate of the population under study. The ap-proach is particularly attractive in instances inwhich classic measures of cellular growth (e.g.,biomass, cell density) are not possible or feasi-ble. For the appropriate implementation of thetechnique, incubation should extend for a mini-mum of 1.0 to 1.5 generations. This derives fromthe fact that data described by equation 1 mustpossess sufficient curvature for the accuratedetermination of k by nonlinear regression tech-niques.The competition for sulfate incorporation into

protein by methionine and cystine can pose aproblem in the interpretation of protein synthe-sis measurements when the organic compoundsare present at moderate levels (ca. 1 ,uM). Me-thionine is especially effective at replacing sul-



http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


fate-sulfur in protein in P. halodurans, withcharacteristics identical to those observed forEscherichia coli (27). The results of competitionwith methionine and cystine also suggest that alarge percentage of the LMW organic sulfur poolis glutathione. The replacement with methionineof part of the residue protein-S normally derivedfrom sulfate would reduce the amount of sulfate-"5S incorporated into this fraction. However, thelower growth rate of P. halodurans with methio-nine as the sole source of sulfur relative tocysteine or glutathione-grown cells indicatesthat the reverse reaction of methionine to cys-teine is unfavorable (7). Therefore, intracellularcysteine and glutathione are most likely synthe-sized from sulfate. Since much of the protein-Sis derived from methionine in its presence, adisproportionately high percentage of the totalsulfate-35S would be in LMW organic-S com-pounds (cysteine plus glutathione), as observed.The results of cystine competition are consistentwith this mechanism, since exogenous cystineshould replace sulfate in intracellular pools ofboth cysteine and the cysteine-containing pep-tide. A detectable decrease in the LMW organic-S derived from sulfate was observed, but theeffect was small. Grown with cystine as the solesource of sulfur, P. halodurans achieves onlyhalf the growth rate obtained with sulfate (7), socystine transport may be too difficult to provideenough sulfur to replace sulfate for metabolismat low cystine concentrations. The absence of adetectable effect of glutathione on the sulfate-35Sdistribution or incorporation into residue proteinindicates that glutathione is not acting as areservoir of cysteine for protein synthesis.The failure to detect sulfur-containing amino

acids in seawater (6, 11, 25, 29, 36), even incoastal waters, means that competition for sul-fate uptake will be negligible in isolated watersamples. However, an investigation of the turn-over times and flux of regenerated LMW organicsulfur compounds into microbial cells is war-ranted.The initial application of sulfate incorporation

into protein, using an organic-rich seawater sys-tem with a well-developed microbial flora, dem-onstrated the value of using an inorganic tracerof bacterial activity. Assimilation of glucose andthat of sulfate were very similar during the firstfew hours of the experiment, but nearly com-plete mineralization of glucose indicated thepreferential utilization of dissolved LMW mate-rial in the early stages of the incubation. Sulfateincorporation into protein was sensitive to thesecondary growth period because it is neverutilized to a significant extent as a percentage ofthe total label and is incorporated in directrelation to growth. The change in distribution of14C after the plateau in total uptake suggests that

the glucose-assimilating portion of the popula-tion also experienced a secondary growth phase,during which time label previously incorporatedinto precursor pools was chased into end prod-ucts by a new source of carbon and energy. Thefish food used in the tank consisted primarily ofpolymeric materials (crude protein and plantfibers), and several hours could be required forthe induction, synthesis, and action of hydrolyt-ic enzymes required for the degradation of pro-teinaceous and cellulosic polymers. The mono-mers thus provided could stimulate continuedgrowth which would not be observed by glucoseassimilation.

In addition to having observed the secondarygrowth phase in the bacterial population fromthe fish tank, the ratio of residue protein-S/totalprotein (Table 1) can be used to calculate theamount of bacterial protein synthesized duringthe incubation period. The operational value,0.93% by weight, indicates the synthesis of 123,ug of protein liter-' during the first hour, a veryhealthy amount of growth. Since protein synthe-sis values for other natural populations are notavailable for comparison, an assumption of 50ocarbon by weight in protein gives a little over 60,ug of protein-C liter-' h-1, the minimum carbonproduction (since carbon is contained in manynonprotein cellular constituents). Values for nat-ural bacterial population growth in a rich coastalinlet in British Columbia (14) determined by cellcounts and thymidine incorporation into DNArange from 0.7 to 70 ,ug of C liter-' day-. Thefish tank population was therefore more than 25times as productive as the most rapid growthobserved in their study.

In summary, the relatively constant sulfurcontent of protein among marine bacteria and itsindependence of growth conditions make sulfateincorporation into protein a reliable measure ofnet bacterial protein production. The major limi-tation to its use in seawater is the high activitynecessary to detect bacterial metabolism (10 to20 ,uCi ml-'). However, the use of appropriateifitration and sample processing methods (8)yields low and reproducible blanks, as well asproviding a more readily interpretable resultthan whole-cell sulfate uptake.

ACKNOWLEDGMENTSThe skillful technical assistance of Steven Molyneaux is

greatly appreciated.This work was supported by National Science Foundation

grants OCE77-12172, OCE79-19178, and OCE79-19264. Addi-tional support to one of us (R.L.C.) was provided by theEducation Department of the Woods Hole OceanographicInstitution.

LITERATURE CITED

1. Abrams, W. R., and J. A. Schiff. 1973. Studies of sulfateutilization by algae. 11. An enzyme-bound intermediate inthe reduction of adenosine-5'-phosphosulfate (APS) by

VOL. 43, 1982


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/


cell-free extracts of wild-type Chlorella and mutantsblocked for sulfate reduction. Arch. Mikrobiol. 94:1-10.

2. Apontoweil, P., and W. Berends. 1975. Glutathione bio-synthesis in Escherichia coli K12. Properties of the en-zymes and regulation. Biochim. Biophys. Acta 399:1-9.

3. Britten, R. J., and F. T. McClure. 1962. The amino acidpool in Escherichia coli. Bacteriol. Rev. 26:292-335.

4. Bradford, M. M. 1976. A rapid and sensitive method forthe quantitation of microgram quantities of protein utiliz-ing the principle of protein-dye binding. Anal. Biochem.72:248-254.

5. Busby, W. F., and A. A. Benson. 1973. Sulfonic acidmetabolism in the diatom Navicula pelliculosa. Plant CellPhysiol. 14:1123-1132.

6. Clark, M. E., G. A. Jackson, and W. J. North. 1972.Dissolved free amino acids in southern California coastalwaters. Limnol. Oceanogr. 17:749-758.

7. Cuhel, R. L., C. D. Taylor, and H. W. Jannasch. 1981.Assimilatory sulfur metabolism in marine microorga-nisms: characteristics and regulation of the sulfate trans-port system in Pseudomonas halodurans and Altero-monas luteo-violaceus. J. Bacteriol. 147:340-349.

7a.Cuhel, R. L., C. D. Taylor, and H. W. Jannasch. 1981.Assimilatory sulfur metabolism in marine microorga-nisms: sulfur metabolism, growth, and protein synthesisof Pseudomonas halodurans and Alteromonas luteo-vio-laceus during sulfate limitation. Arch. Microbiol. 130:1-7.

7b.Cuhel, R. L., C. D. Taylor, and H. W. Jannasch. 1981.Assimilatory sulfur metabolism in marine microorga-nisms: sulfur metabolism, protein synthesis, and growthof Pseudomonas halodurans and Alteromonas luteo-vio-laceus during unperturbed batch growth. Arch. Micro-biol. 130:8-13.

8. Cuhel, R. L., C. D. Taylor, and H. W. Jannasch. 1981.Assimilatory sulfur metabolism in marine microorga-nisms: sulfur metabolism, protein synthesis, and growthof Alteromonas lutes-violaceus and Pseudomonas halo-durans during perturbed batch growth. Appl. Environ.Microbiol. 43:151-159.

9. Daley, R. J., and J. E. Hobble. 1975. Direct counts ofaquatic bacteria by a modified epifluorescence technique.Limnol. Oceanogr. 20:875-882.

10. Datko, A. H., S. H. Mudd, J. Giovanelli, and P. K. Macni-col. 1978. Sulfur-containing compounds in Lemna perpu-silla 6746 grown at a range of sulfate concentrations. PlantPhysiol. 62:629-635.

11. Dawson, R., and G. Gocke. 1978. Heterotrophic activity incomparison to the free amino acid concentrations in Balticsea water samples. Oceanologica Acta 1:45-54.

12. Eppley, R. W., R. W. Holmes, and J. D. H. Strickland.1967. Sinking rates of marine phytoplankton measuredwith a fluorometer. J. Exp. Mar. Biol. Ecol. 1:191-208.

13. Fahey, R. C., W. C. Brown, W. B. Adams, and M. B.Worsham. 1978. Occurrence of glutathione in bacteria. J.Bacteriol. 133:1126-1129.

14. Fuhrman, J. A., and F. Azam. 1980. Bacterioplanktonsecondary production estimates for coastal waters ofBritish Columbia, Antarctica, and California. Appl. Envi-ron. Microbiol. 39:1085-1095.

15. Glovanelli, J., L. D. Owens, and S. H. Mudd. 1973. -Cystathionase: in vivo inactivation by rhizobitoxine andthe role of the enzyme in methionine biosynthesis in cornseedlings. Plant Physiol. 51:492-503.

16. Herbert, D. 1961. The chemical composition of micro-organisms as a function of their environment, p. 391-416.In G. G. Meynell and H. Gooder (ed.), Microbial reactionto environment (llth Symposium of the Society for Gen-

eral Microbiology). University Press, Cambridge.17. Holmquist, R. 1978. Evaluation of compositional nonran-

domness in proteins. J. Mol. Evol. 11:349-360.18. Jordan, M. J., and B. J. Peterson. 1978. Sulfate uptake as

a measure of bacterial production. Limnol. Oceanogr.23:146-150.

19. Jukes, T. H., R. Holmquist, and H. Moise. 1975. Aminoacid composition of proteins: selection against the geneticcode. Science 189:50-51.

20. Karl, D. M. 1979. Measurement of microbial activity andgrowth in the ocean by rates of stable ribonucleic acidsynthesis. Appl. Environ. Microbiol. 38:850-860.

21. Kredich, N. M. 1971. Regulation of L-cysteine biosynthe-sis in Salmonella typhimurium. I. Effects of growth onvarying sulfur sources and O-acetyl-L-serine on geneexpression. J. Biol. Chem. 246:3474-3484.

22. Kredich, N. M., M. A. Becker, and G. M. Tomkins. 1969.Purification and characterization of cysteine synthetase, abifunctional protein complex, from Salmonella typhimur-ium. J..Biol. Chem. 244:2428-2439.

23. Law, A. T., and D. K. Button. 1977. Multiple-carbon-source-limited growth kinetics of a marine coryneformbacterium. J. Bacteriol. 129:115-123.

24. Lewin, J., and W. F. Busby, Jr. 1967. The sulphaterequirements of some unicellular marine algae. Phycolo-gia 6:211-217.

25. Lindroth, P., and K. Mopper. 1979. High pressure liquidchromatographic determination of subpicomole amountsof amino acids by precolumn fluorescence derivatizationwith o-phthaldialdehyde. Anal. Chem. 51:1667-1674.

26. Parsons, T. R., and J. D. H. Strickland. 1962. On theproduction of particulate organic carbon by heterotrophicprocesses in seawater. Deep Sea Res. 8:211-222.

27. Roberts, R. B., P. H. Abelson, D. B. Cowie, E. T. Bolton,and R. J. Britten. 1963. Studies of biosynthesis in Esche-richia coli. Carnegie Inst. Washington PubI. 607.

28. Rosenbauer, R. J., J. L. Bischoff, and W. E. Seyfried.1979. Determination of sulfate in seawater and naturalbrines by "3Ba and membrane dialysis. Limnol. Ocean-ogr. 24:393-3%.

29. Schell, D. M. 1974. Uptake and regeneration offree aminoacids in marine waters of southeast Alaska. Limnol.Oceanogr. 19:260-270.

30. Schiff, J. A., and R. C. Hodson. 1970. Pathways of sulfatereduction in algae. Ann. N.Y. Acad. Sci. 175:555-576.

31. Schmidt, A., W. R. Abrams, and J. A. Schiff. 1974. Re-duction of adenosine-5'-phosphosulfate to cysteine inextracts from Chlorella and mutants blocked for sulfatereduction. Eur. J. Biochem. 47:423-434.

32. Taylor, C. D. 1979. Growth of a bacterium under a high-pressure oxyhelium atmosphere. Appl. Environ. Micro-biol. 37:42-49.

33. Taylor, C. D., and H. W. Jannasch. 1976. Subsamplingtechnique for measuring growth of bacterial cultures un-der high hydrostatic pressure. Appl. Environ. Microbiol.32:355-359.

34. Tempest, D. W., J. L. Meers, and C. M. Brown. 1970.Influence of the environment on the content and composi-tion of microbial free amino acid pools. J. Gen. Microbiol.64:171-185.

35. Tindall, D. R., J. H. Yopp, W. E. Schmid, and D. M.Miller. 1977. Protein and amino acid composition of theobligate halophile, Aphanothece halophytica (Cyano-phyta). J. Phycol. 13:127-133.

36. WillIams, P. J. LeB., T. Berman, and 0. Holm-Hansen.1976. Amino acid uptake and respiration by marine het-erotrophs. Mar. Biol. 35:41-47.



http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

assimilatory sulfur metabolism microorganisms ... · protein synthesis in unenriched...

Documents