Alkalophilic BACTERIA
Alkalophilic BACTERIA
Alkalophilic BACTERIA
REVIEWS Further
Quick links to online content
Annu. Rev. Microbial. 1989. 43:435-{;3
Copyright © 1'989 by Annual Reviews Inc. All rights reserved
ALKl\LOPHILIC BACTERIA
CONTENTS
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
BIODIVER SITY OF ALKALOPHILE S . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436
COMMON PROPERTIES OF EXTREME A LKALOPHILES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440
Sodium Cycle Facilitating pH Homeostasis and Solute Uptake . . . . . . . . . . . . . . . . . . . . . . '" 440
Requirement of an Electrochemical Sodium Gradient for Motility . . . . . . . . . . . . . . . . . . . . . . 443
Use of Energy From a Proton-Extruding Respiratory Chain and a Proton-
Translocating F/-Fo ATPase for Oxidative Phosphorylation . . . . . . . . . . . . . . . 444
Properties of Cell Components That are Exposed to the Exterior . . . . . . . . . . . . . . . . . . . . . . 451
BIOTECHNOLOGY OF ALKALOPHILE S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452
Enzymes and Other A lkalophile Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452
GENETIC E NGINEERING USING ALKALOPHILE GENE S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 456
PRO SPECTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 457
INTRODUCTION
435
0066-4227/89/1001-0435$02.00
436 KRULWICH & GUFFANTI
which can grow well in the neutral pH range, and obligate alkalophiles , which
show optimal growth above pH 10.0 but cannot grow below pH 8. 5-9.0.
The facultative alkalophiles are of particular interest because they offer an
experimental opportunity to use pH-conditional mutants to investigate the
structural and functional features that enable a cell to thrive at very high pH.
Several properties are shared by all the extreme alkalophiles studied to date.
These include the composition of membrane lipids and the membrane lipid!
protein ratio, very high levels of respiratory-chain components in the mem
brane, a generally more acidic amino acid composition of proteins that are
exposed to or excreted into the external milieu, and a Na + cycle that facili
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
tates solute uptake and pH homeostasis (88). Any or all of these properties
Access provided by 103.206.136.102 on 08/12/21. For personal use only.
BIODIVERSITY OF ALKALOPHILES
far thermotolerant alkalophiles have been used for production of enzymes, but
a full characterization of a highly thermophilic and alkalophilic strain has
been repOlted only recently. In 1978, Kitada et al (75) described a thermophil
ic alkalophile that they designated Bacillus strain IC and identified as a strain
of Bacillus licheniformis. Bacillus strain IC grew optimally at 58°C and grew
well between pH 8.0 and 10.0, with some growth at neutral pH. One
intracellular enzyme, enolase, proved to be stable up to 60°C and was most
active in the neutral pH range. Extracellular enzymes have not yet been
characterized. A less alkalophilic species, Methanobacterium thermoalca
liphilum (7), appears to be the first thermophilic methanogen exhibiting good
growth in the high pH range. It grows up to pH 9.0-9 . 5 , with optimal growth
between pH 7 . 5 and 8 . 5 . This alkaline-tolerant anaerobe shows optimal
growth at 60°C. At the other end of the temperature spectrum, several as yet
unidentifie:d bacterial strains that grow well at pH 10.0 and at temperatures as
low as O°C have been isolated (72).
Marine environments abound in alkaline-tolerant bacteria and will general
ly yield small numbers of true alkalophiles from suitable enrichment cultures.
These species usually belong to the major alkalophile genus, i.e. Bacillus. By
contrast, the large number of gram-positive and gram-negative alkaline
tolerant and halotolerant species fall into many genera. Most recently, an
alkaline-tolerant coryneform bacterium was isolated from seawater ( 1 09). Its
growth was optimal between 0.5 and 2.0 M NaCl or KCI and occurred over a
broad pH range, from below 7.0 up to 9 . 5 . Interestingly, although most
extreme alkalophiles are Bacillus species, extensive studies of highly alkaline
saline lakes (28 , 142) have yielded only one possible member of this group.
There may be special biological problems associated with the combined
stresses of halophilicity and alkalophily.
Most of the truly alkalophilic microorganisms have either been isolated
from specific, enriched environments such as indigo dye balls ( 1 30), potato
processing-plant effluents ( 1 3) , or alkaline lakes (29, 1 24, 1 32) or have been
isolated upon suitable enrichment culturing of soil. The eubacterial alka-
..,.
w
Table 1 Diversity of alkalophilic bacteria 00
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
Eubacteria c::
Gram-negative photo-
autotrophs
�
-
n
Spirulina sp. 8.0-11.0 Alkaline soda lakes, Rift Val- Halotolerant cyanobacterium 29 :I:
ley, pH 10.5 ?:o
Synechococcus sp. 6.5-10.0, optimum Yellowstone, 55°C, pH 5.5 Alkali-tolerant, halotolerant 61 0
c::
pH 8.0 cyanobacterium
Ectothiorhodospira sp. 8.0-10.0 Mud from alkaline salt lakes, Anoxygenic purple sulfur bacterium 27
�
>
pH 11.0 Z
::l
Gram-negative
nonendosporeformers
Aeromonas sp. 7.0-10.0 Soil Facultative anaerobe 53
Flavobacterium sp. 7.0-1 1.2 Soil Facultative anaerobe 56
Vibrio alginolyticus 6.0-9.2 Seawater Halotolerant, alkali-tolerant faculta- 1 35
tive anaerobe
Gram-positive
nonendosporeformers
Exiguobacterium sp. 7.0-11.5, optimum Potato-processing effluent Facultative anaerobe 13
8.5-9.5
Actinomyces sp. 8.0-1 1.5, optimum Soil Heterotrophic facultative 105
9.0-9.5 anaerobe
Streptococcus sp. 5.0-11.0, optimum Alkaline potato processing Fermentative facultative 26
8.0-9.0 effluent anaerobe
Coryneform bacteria 6.6-9.5 Seawater Halotolerant facultative 109
anaerobes
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
Gram-positive
endosporeformers
Access provided by 103.206.136.102 on 08/12/21. For personal use only.
Archaebacteria �
�
Aerobes
Natronobacterium
Natronococcus sp.
sp. optimum 9.0-10.0 Alkaline saline lakes Halophilic (3 M NaCI) 132
[;
:i!
Anaerobes E
n
Methanobacterium 6.5-10.0 Biogas plant, optimum Thermophilic autotroph 7 t:::tl
thermoalcaliphilum pH 7.5-&.5 :.>
q
trl
�
:.>
.j>.
W
\0
440 KRULWICH & GUFFANTI
lophiles isolated from soil are usually, but not always, Bacillus species.
Facultative alkalophiles have recently been isolated from soil and assigned to
the genus Flavobacterium (56). A group of these strains grew at pH 7. 1 - 1 1 . 2,
with better initial growth at pH 10.2 than at pH 7 . 2; a-galactosidase produc
tion was also better at very high pH. Another category of alkalophilic soil
organisms, the alkalophilic actinomycetes, has recently been reviewed ( 105) .
Several isolates have been shown to b e obligate alkalophiles, exhibiting
excellent growth up to pH 1 1.0. Notably, the alkalophilic actinomycetes and
many of the alkalophilic Bacillus species have been isolated from soils that
are not particularly alkaline. Although the organisms are more prevalent in
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
alkaline soils ( 1 29), true alkalophiles have been isolated even from acidic
Access provided by 103.206.136.102 on 08/12/21. For personal use only.
soils. Presumably there are microenvironments that allow the growth of even
extreme, obligate alkalophiles.
+ +
Na /H ANTIPORTER AND OTHER POSSIBLE MECHANISMS FOR NET H+
values at the upper pH limit for growth were near zero (-1 5 mV i n that study,
and higher when malate was included in the buffer). How could the antiport
be energized to achieve the proton gradient found if the ajj,wlF were as low
as it seemed to be? At ajj,H+IF 0, an electrogenic antiporter could not be the
=
sole underlying mechanism. The real values, however, in energized cells (and
even cells suspended in buffer) are never zero and could be achieved at very
reasonable: H+/Na+ stoichiometries.
The cmcial involvement of the Na+ /H+ antiporter is supported by the
isolation of nonalkalophilic mutant strains that can no longer grow well above
pH 9.0 and have lost Na+ /H+ antiporter activity (79, 85 , 89). Also important
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
potential in cells, vesicles, and proteoliposomes (85, 89, 91). When the same
methods were used at somewhat lower pH values, many aerobic eubacteria
seemed to possess a comparable activity, e.g. Vibrio alginolyticus (110),
Escherichia coli (5, 120), and even the archaebacterium Halobacterium
halobium (95). It is probable that in many organisms the antiporter is required
for growth under moderately alkaline conditions, such as at pH 8.5-9. 0. This
idea was proposed some time ago for E. coli (114, 146), and compelling
recent evidence in its favor has emerged (60, 103). Less clear is whether
there are other ion fluxes or mechanisms that have an important role in
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
not very different from that for E. coli growth, a K+/H+ antiporter has a
role in maintaining a constant pHin at high pH, even though a Na+/H+ anti
porter is still apparently required for major tuning. Koyama et al (83,
84) have proposed, by contrast, that fluxes of K+ are involved in alkaliniz
ing rather than in acidifying the cytoplasm of extremely alkalophilic bacilli.
Perhaps K+ fluxes are more relevant, then, to pH homeostasis at the
low end of the pH range for growth. The studies of Koyama et al (84) in
dicated that K+ uptake was electrogenic at neutral pH and electroneutral
at high pH. They correlated this apparent difference with a greatly en
hanced permeability of the alkalophile membrane to K+ at neutral ver
sus high pH. It is notable that Matsukura & Imae (101) found that the
permeability to K + and the K + -dependent diminution of the 11r/J of several
different alkalophilic bacilli at high pH was greatly enhanced when Na+
was omitted. This effect also seemed pronounced at pH 7. 5 (101) and
may thus underlie the observations of Koyama et al (84). The mecha
nism of the Na+ effect on the permeability to K+ is unknown, but it will be
of interest to determine whether it is a global effect on the bulk membrane,
a Na + -dependent exchange phenomenon, or some other more specific
effect.
Finally, with respect to activities that catalyze or facilitate net proton
accumulation by extreme alkalophiles, Koyama et al (82) recently reported a
facultatively anaerobic alkalophile that maintains a cytoplasmic pH below the
external pH value even though (a) the total l1[.tw is positive, since the I1pH is
reversed and there is either no 11r/J or a very small 11r/J, positive out; (b) the
I1pH is only partially Na+ dependent; (c) the organism contains no
cytochromes; and (d) the I1pH is not dissipated by protonophores. Koyama et
al (82) proposed that a Donnan potential allows these bacteria to acidify the
interior even in the absence of Na+. It is not clear whether the N a+-dependent
component might result from an antiport activity, perhaps energized suf
ficiently by the very small Ar/J, which could arise, for example, by ATP
hydrolysis.
ALKALOPHILIC BACTERIA 443
bacteria in symport with Na+ began with findings in the mid-1 970s (74, 80).
Access provided by 103.206.136.102 on 08/12/21. For personal use only.
and motility was highly susceptible to some amiloride analogs that are potent
Na+ channel inhibitors (126).
Thus far the motility of the alkalophilic bacilli has appeared to be de
pendent exclusively upon Na+, with neither substitution by Li+ at high pH
(44) nor any energization by a dpH at near-neutral pH (128). To date, even
facultative strains that grow over a broad pH range have shown this Na+
specificity for both motility and Na+-coupled solute symporters (128). This
rigid specificity is unexpected, because solute symporters in other organisms
at least sometimes accommodate Na+, Li +, or H+ as the coupling ion (139).
Some nonalkalophilic mutants of alkalophiles may use H+ as the coupling ion
for solute symport (88). In addition, a growing literature exists on the
modification of coupling-ion specificity for Na + -coupled porters upon various
mutational alterations (8, 68, 112).
Although Na+ appears to be an obligatory component of the energization
system for alkalophile motility, a chemical gradient of Na + alone may not
energize motility in the absence of a dl/l in untethered cells. Only for the
modestly alkaline-tolerant marine vibrios has it been possible to demonstrate
motility upon imposition of an artificial apNa+ upon freely swimming cells.
Cells of V. alginolyticus were rendered immotile by treatment with arsenate
and 2-n-heptyl-4-hydroxyquinoline-N-oxide (HQNO) before imposition of
the Na+ gradient (16); however, because the levels of HQNO were IO-fold
lower than those shown to abolish the dl/l totally in the same study, it is not
completely clear whether a residual dl/l was present during the imposition of
the dpNa+.
A
H+(2-3/ATP)
C.pH>+120mV
c.'I' =-170mV Solutes
+ C.}lH·�- 50mV
C.Gp =-499mV
No'
SOhJleS
isms) does not result in ATP synthesis (30, 33, 34). Although the calculated
magnitudes of the chemiosmotic driving force derived from respiration, an
Access provided by 103.206.136.102 on 08/12/21. For personal use only.
(D. B . Hicks, unpublished) the FIFo-ATPase from B . firmus RAB has been
Access provided by 103.206.136.102 on 08/12/21. For personal use only.
OF4, the amount of ATP synthesized was linear with the magnitude of the
imposed j�pH over the range studied (34). However, the total amount of ATP
synthesizl�d at steady state upon imposition of a dpH of one unit was much
smaller than that observed in respiring whole cells whose d{Lw/F of 50 mV
-
is a smaller bulk driving force (34). Were the ATPase to translocate a large
number of protons per ATP molecule synthesize, it should have been capable
of doing so in the experiments with the imposed dpH. Even more strikingly,
if the extreme alkalophiles used a high H+:ATP stoichiometry to resolve the
problem of low-bulk d{Lw, then a high stoichiometry should have allowed
the organisms to synthesize ATP at pH 9.0 upon energization by a K+
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
diffusion potential.
Access provided by 103.206.136.102 on 08/12/21. For personal use only.
obstructive proteins that would simply get in the way; the fluidity of the
membrane or other lipid characteristics that would influence protein diffusion
rates; and the juxtaposition of the requisite residues or the proton leak into the
bulk.
This type of model accounts for the following observations about the
extreme alkalophiles. ATP synthesis at low /1jLwlF, but never at /1jLH+IP 0 =
would be consistent with a model in which the bulk force resisted outward
leakage of protons but was not the primary, direct energetic force . The
primary force would be provided by directly moving protons that are not
measured when the bulk /1jLH+ measurements are made. The localized flow
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
Figure 2 Schematic representation of A hypothesis that accounts for the bioenergetic properties
of extreme alkalophiles. Energy coupling at very high pH is proposed to occur largely (perhaps
even exclusively) through actual collisions between the proton-binding polypeptides of the
respiratory chain and proton-binding residues of the FIFo-ATPase. The frequency of those
collisions is high, and the path of proton flow to and from the exterior is of relatively high
resistance, so that localized proton flow is facilitated. The high collision frequency would be
favored by the very high concentration of respiratory-chain components (hatched areas) that is
characteristic of these organisms, by a somewhat high lipid/protein ratio in the membrane [i.e.
fewer obstructing proteins (solid circles) than in some other coupling membranes], and by high
membrane fluidity to allow the protein diffusion that would be required for collisions to occur.
Some components of the membrane might also enhance proton binding or favor the formation of
domains within the membrane lipids such that respiratory-chain components and FIFo-ATPase
are trapped together. This system would work only with natural H+ pumps. It may be dependent
upon an especially high concentration of proton pumps and sinks (or H+ -binding segments
thereof). Other aspects of the model are discussed in the text.
ALKALOPHILIC BACTERIA 451
the membJrane lipids and the possession of polar isoprenoid lipids that may
Access provided by 103.206.136.102 on 08/12/21. For personal use only.
allow domain formation are related to energy coupling (12, 91). Importantly,
the model should be testable through genetic and biochemical approaches.
counterpart (31), and a cytochrome c that may well have an exposed localiza
tion is an acidic protein (14).
BIOTECHNOLOGY OF ALKALOPHILES
C-terminal part of the enzyme is not needed for activity. The Bacillus strain
1 1 39 cellulase showed strong nucleotide sequence homology in the N
tenninal re:gion toward the other two smaller cellulases from Bacillus strain
N-4.
elastase from alkalophilic Bacillus strain Ya-B ( 1 37) had a pI of 10.6 and
Access provided by 103.206.136.102 on 08/12/21. For personal use only.
rather than 145 kd. It was speculated that this difference might be due to
glycosylation occurring in Aeromonas spp. but not in E. coli. More recently,
Kato et al (65) cloned the xylanase L gene into E. coli by using a pEAP37
or pEAP2 excretion vector and achieved extracellular enzyme production
(67) .
Alkalophilic Bacillus strain C- I 25 produces two xylanases , xylanase A and
xylanase N, of 43 and 1 6 kd, respectively (46 , 50) . Xylanase N was most
active at pH 6.0-7 .0, whereas xylanase A was most active at pH 6.0-10.0;
the latter was active even up to pH 12.0, with a temperature optimum of 70°C.
When cloned into E. coli, xylanase A was produced extracellularly (38 , 48,
49) and exhibited similar properties to the enzyme from Bacillus strain C- I 25
(46, 47) , Ibut the E. coli borne enzyme was slightly less stable at low pH
-
Using glycine
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
to "loosen" the cell surface components , Ikura & Horikoshi (55) caused the
Access provided by 103.206.136.102 on 08/12/21. For personal use only.
plasmid containing the human immunoglobulin G-Fc region and kil gene. E .
coli carrying the plasmid excreted about 40% of the human immunoglobulin
G-Fc region into the culture medium (76); the rest remained in the periplasm
and cytoplasm. The same protocol was used to introduce the gene for human
growth hormone (hGH) into E. coli (64); 55% of the total hGH was excreted
into the medium and 42% into the periplasm. The mature hGH was processed
normally and exhibited authentic hGH activity. Thus, activation of the kil
gene by an alkalophilic promoter opens up interesting opportunities for
establishing excretion of appropriate gene products by E. coli.
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
PROSPECTS
Access provided by 103.206.136.102 on 08/12/21. For personal use only.
Literature Cited
Bacillus sp. strain 1 1 39. J. Gen. Micro acid transport in an obligately alkalo
bioi. 132:2329-35 philic bacterium. J. Bioi. Chem.
24. Fukumori, F . , Ohnishi, F. K . , Kudo, 253:708- 1 5
T . , Ho:rikoshi, K. 1987 . Tandem loca 36. Hackenbrock, C . R . , Chazotte, B . ,
tion of the cellulase genes on the Gupte, S . S . 1986. The random collision
chromosome of Bacillus sp. strain N-4. model and a critical assessment of diffu
FEMS Microbiol. Lett. 48:65-68 sion and collision in mitochondrial elec
25. Fukumori , F. , Sashihara, N . , Kudo, T . , tron transport. J. Bioenerg. Biomembr.
Horikoshi, K . 1986. Nucleotide se 1 8:33 1-68
quences of two cellulase genes from 37. Hamamoto, T. , Honda, H . , Kudo, T . ,
alkaloplhilic Bacillus sp. strain N-4 and Horikoshi, K . 1987. Nucleotide se
their sitrong homology. 1. Bacteriol. quence of the xylanase A gene of
168:479-85 alkalophilic Bacillus sp. strain C- 1 25 .
26. Graham, A. F . , Lund, B . M . 1983. The Agric. Bioi. Chem. 5 1 :953-55
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
C-125 by Escherichia coli carrying kalophilic Bacillus sp. strain No. 38-2.
pCX3 1 1 . Syst. Appl. Microbiol. 8 : 152- J. Gen. Microbiol. 1 34:97-105
57 63. Kaneko, T . , Kato, T . , Nakamura, N . ,
49. Honda, H. , Kudo, T . , Horikoshi, K. Horikoshi, K. 1987. Spectrophotometric
1986. Extracellular production of alka determination of cyclization activity of
line xylanase of aJkalophilic Bacillus sp. l3-cyclodextrin--forming cyclomaltodex
by Escherichia coli carrying pCX3 1 1 . J. trin glucanotransferase. J. Jpn . Soc.
Ferment. Teehnol. 64:373-77 Starch Sci. 34:45-48
50. Honda, H . , Kudo, T . , !kura, Y . , Hori 64. Kato, c . , Kobayashi, T . , Kudo, T . ,
koshi, K. 1985. Two types of xylanases Furusato, T., Murakami, Y . , et at.
of alkalophilic Bacillus sp. No. C- 1 25 . 1987. Construction of an excretion vec
Can. J. Microbiol. 3 1 :538-42 tor and extracellular production of hu
51. Horikoshi, K. 1985. Problems of bio man growth hormone from Escherichia
technology and their solution-Establish coli. Gene 54: 197-202
Annu. Rev. Microbiol. 1989.43:435-463. Downloaded from www.annualreviews.org
ment and application of excretion sys 65. Kato, C . , Kobayashi, T. , Kudo, T.,
tems. Chern. Eeon. Eng. Rev. 1 7 : 1 2- Horikoshi, K. 1986. Construction of an
Access provided by 103.206.136.102 on 08/12/21. For personal use only.