Anaerobic Degradation of Linoleic (Clsa), Oleic (C18:i) and Stearic (Clsa) Acids and their Inhibitory Effects on Acidogens, Acetogens and Methanogens Jerald David Anthony Lalman A thesis submitted in conformity with the requirements For the degree of Doctor of Philosophy Graduate Department of Civil Engineering University of Toronto OCopyright 2000 by Jerald Anthony David Lalrnan
202
Embed
Anaerobic Degradation of Linoleic Oleic (C18:i) and Stearic Acids … · 2020-04-07 · Anaerobic Degradation of Linoleic (Cl82), Oleie (Ci8,i) and Stearic (Cis:o) Acids and their
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Anaerobic Degradation of Linoleic (Clsa), Oleic (C18:i) and Stearic (Clsa) Acids and their Inhibitory Effects on Acidogens, Acetogens and Methanogens
Jerald David Anthony Lalman
A thesis submitted in conformity with the requirements For the degree of Doctor of Philosophy
Graduate Department of Civil Engineering University of Toronto
OCopyright 2000 by Jerald Anthony David Lalrnan
National Library Bibliothèque nationale 1*1 of Canada du Canada
Acquisitions and Acquisitions et Bibliographie Services services bibliographiques 395 Wellington Street 395. rue Wdlingtm OttawaON K 1 A W OttawaON K I A W Canada Canada
The author has granted a non- exclusive licence allowing the National Library of Canada to reproduce, loan, distribute or sel1 copies of this thesis in microfom, paper or electronic formats.
The author retains ownership of the copyright in this thesis. Neither the thesis nor substantial extracts fiom it may be p ~ t e d or othewise reproduced without the author's permission.
L'auteur a accordé une licence non exclusive permettant à la Bibliothèque nationale du Canada de reproduire, prêter, distribuer ou vendre des copies de cette thèse sous la forme de microfiche/fiim, de reproduction sur papier ou sur format électronique.
L'auteur conserve la propriété du droit d'auteur qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation.
Anaerobic Degradation of Linoleic (Cl82), Oleie (Ci8,i) and Stearic (Cis:o) Acids and their Inhibitory Effects on Acidogens, Acetogens and Methanogens
Jerald David Anthony Lalman Doctor of Philosophy
Department of Civil Engineering University of Toronto
2000
Abstract
Effiuents fiom many food processing industries contain fats and oils in addition to
carbohydrates and proteins. Long chain fatty acids (LCFAs), a hydrolysis byproduct of
fats and oils, are difticult to degrade and are inhibitory to anaerobic organisms. These
acids are degraded via f3-oxidation but the compound initiating the mechanism has not
been clearly identified. Although LCFAs inhibit aceticlastic methanogenesis their effects
on acidogenesis, acetogenesis and hydrogenotrophic methanogenesis have not been well
studied.
This study assessed the degradability of linoieic (C 8:?), oleic (Ci 1) and steark (C
acids and determined their inhibitory effects on anaerobic organisms in 160 mL s e m
bottles. Degradation and inhibition studies were conducted using 10, 30, 50 and 100
r n g ~ - ' LCFA. Inhibition studies using glucose, butyrate and acetate (each at 100 rng-~- ' )
and hydrogen (10.1 kPa) investigated the effects of the three LCFAs on acidogens,
acetogens, acet iclastic methanogens and hydrogenotrophic methanogens.
Unsaturated C 18 LCFAs were degraded to shorter chah LCFAs however, no LCFA
byproducts were detected as intermediates during the degradation of stearic (C18:o) acid.
Palmitic and myristic (CI~:o) acids were produced fiom linoleic CI^:^) acid at al1
concentrations examined and in cultures receiving more than 10 rng-~- l oleic (Ci8:l) acid.
In cultures receiving 100 rng-~' ' Linoleic (C 1 a2) acid both oleic (C 1 !) and palrnitoleic
(C 16:1) acids were detected.
Acidogenesis was af5ected by the presence of LCFAs and synergistic inhibitory
effects of al1 three acids on acetogenesis were observed. Hydrogenotrophic
methanogenic inhibition was observed and aceticlastic methanogens were inhibited at al1
LCFA concentrations examined. in cultures fed with linoleic (C 1 g:2) and oleic (C 18: I )
acids, inhibition of acetate methanogenesis was concentration dependent but for cultures
receiving stearic (C ls:o) acid, the effect was independent of concentration.
In comparison to stearic CI^:^) acid, iinoleic (Ci8:2) and oieic (Cla:~) acids were
degraded faster. Therefore, the design of a full-scaie system will depend on the SRT for
the more slowly degrading LCFA. LCFAs affected glucose and butyrate degradation.
Hence, in effluents containing carbohydrate and LCFAs mixtures, the degradation of
carbohydrate monomers wil be afTected. In comparison to oieic (Cls:~) acid, greater
aceticlastic inhibition was observed for cultures receiving linoleic (Clg:2) acid. Thus, it is
recornmended a two-stage process, acetogenic followed by rnethanogenic, be used to
minimize the inhibition.
... I I I
Acknowledgemen ts
This w-ork is dedicated to my mother. Rosaline Lalman, who passed away while
n ~ i t i n g this thesis. My father. Samuel Lalman. has been a great inspiration throughout
my life and 1 n-ish to thank him for instilling into me the meaning of accomplishrnent. 1
would also like to give my special appreciation to Nin for her precious suggestions. her
stronp encouragement at times of difficulty. love and understanding throughout this
u-hole process. Also. thanks to my brother. Edmund Lalman and my sister. Maria
Crutchley for sharing these troubled times in our lives.
I wish to espress my sincere appreciation to my advisor. Dr. David Bagley for his
intelligent supervision. constructive criticisms. inspiration and friendship. My sincere
appreciation estrnds to m). other cornmittee members Dr. Grant Allen. Dr. Don Kirk and
Dr. Brrnr Slecp. n-hose guidance. assistance and friendship are also invaluable. 1 am also
~ratrsful to Dr. Mary Jans Philips and Durga Prasad for their encouragement and ad~vicr. C
1 n.ould like to th& Rajesh Seth. Russell D'Souza and Yale Zheng for sharing their
kno\vledgs and helpful comments.
Financial assistance was providsd by the University of Toronto and the Ontario
Ministr).. of Energy Science and Technology: Singapore-Ontario joint research
programme.
Finall>-. 1 would like to t h d the Department of Civil Engineering for their support
2.3 Anaerobic Reactor Technologies Used to Treat Emuents Containing Long Chain Fatty Acids 2.3.1 Introduction 2.3 - 2 LOW Rate Treatment 2.3.3 High Rate Treatment
2.4 Long Chain Fatty Acids 2.4.1 Sources and Treatment 2.4.2 Composition and Structure 2.4.3 Biodegradation of LCFAs 2.4.4 Effects of Hydrogen and Volatile Fatty Acid
A . Hydroger~ B. 1,klafile Fatp Acidr
Page No.
. . 11
5
sii
2.4.5 Inhibitory Effects A . Effecrs on Membrane Fmcîion B. Effecrs on Anaerobic Orgmrisms
2.4.6 Factors AfFecting LCFA Degradation and Inhibitory Eftècts A. Szi bstra~e Molemlar Sb-uctlrre and Conceritratiorr B. Tentperarure Eflects C. Soltrbilzty Efects D. pH EfJecrs E. Coslibsirare ami fiermocjtraniic Ef/ecrs
Some hydrolytic reactions and fiee energies 10 Some acidogenic reactions and fiee energies 13 Some acetogenic reactions and fiee energies 14 Some methanogenic reactions and fiee energies 15 Design parameters reported for anaerobic reactors treating effluents containing fats and oils Selected properties of linoleic (C ~ g : ~ ) , oleic (C 18: 1 ) and steark (Clg:~) acids Main edible oil categories EdibIe oils containing Iinoleic (Cl*:?) acid EdibIe oils containing oleic (CI 8: 1 ) acid Edible oils containing stearic (C18:o) acid Hydrogenation of linoleic (ClgI2) acid by rumen bacteria LCFA oxidation reactions Organisms associated with LCFA degradation Fatty acids aqueous solubilities and dissociation constants Free energy changes for C 18 LCFAs oxidation to acetate LCF.4s Gibbs free energy of formation values Change in fiee energy of some relevant j3-oxidation and methanogenic reactions 58 Experimental design matrix 60 Linoleic (C I S : ~ ) , oleic and stearic (C I R O ) acids degradation studies 6 1 Acidogenic inhibition studies conducted with linoleic (CIH:~) , oleic (Cl8:l) and stearic (&O) acids Acetogenic inhibition studies conducted with linoleic (C 1 8 : ~ ) . oleic (Ci8:1) and stearic (C lx:o) acids Aceticlastic rnethanogenic inhibition studies conducted with linoleic (C18:I), oieic (C18:I) and steark (Clrt:~) acids Hydrogenotrophic methanogenic inhibition studies conducted with linoieic (C 18:2), oleic (Clg:l) and stearic (Ci8:o) acids LA/O.LVSA interaction studies Basal medium characteristics Dispersing agent solubilities Maximum concentration of byproducts formed during linoleic ( C l 4 acid degradation Initial acetate degradation rates for varying linoleic (C 1 8 : ~ ) acid concentrations First order rate constants for hydrogen removal 1 day afier adding LA First order rate constants for hydrogen removal 18 days af-ler adding LA
Table 5.5 :
Table 6.1 :
Table 6.2:
Table 7.1 : Table 7.2:
Table 7.3:
Table S . 1 :
Table 8.2: Table 8.3: Table 8.4: Table 8.5:
Table 9.1 :
Table 9.2:
Table 9.3:
Table -4.1 : Table A.2:
Table -4.3 : Table A.3: Table AS: Table B. 1 :
First order rate constants for hydrogen removal 35 days afier adding LA Initial acetate degradation rates for varying oleic ( C I R : ~ ) acid concentrations First order rate constants for hydrogen removal in the presence of oleic (C 18: 1) acid Stearic (C1g:O) acid degradation rates Acetate degradation rates for varying stearic (Clg:o) acid concentrations First order rate constants for hydrogen removal in the presence of stearic (Cig:~) acid Byproducts detected in cultures receiving 100 rng .~ ' ' LCFA Byproducts detected in cultures receiving linoleic (C18:2) acid Byproducts detected in cultures receiving oleic (Cl s: 1) acid Initial LCFA degradation rates Free energy values for fi-oxidation of linoleate (&:2), oleate (Cls: ,) and stearate (C ,s:o) to palmitate (C 16:0) Glucose degradation rates for individual and mixed LCFA su bstrates Butyrate degradation rates for i ndividual and mixed LCFA substrates First order rate constants for hydrogen removal in the presence of individual and mixed LCFAs Constants -4 and B for AG; calculations Calculation of free energy of formation for gaseous LCFAs Estimation of constants for Antoine's equation Estimation of Henry's constant Estimation of ionized LCFA free energy of formation Mass balance calculations for cultures receiving 100 mg^“ Linoleic acid at time 17.3 days
List o f Figures
Figure 2.1 : Anaerobic conversion processes Fisure 2.2: Structure of a polysaccharide Figure 2.3: Structure of a tripeptide Figure 2.4: Structure of a triacylglycerol Figure 2.5: Operating HRT and COD ranges for aerobic and
anaerobic biological treatment technologies Figure 2.6: Anaerobic treatment process classification Figure 2.7: The anaerobic contact process flow schematic Figure2.8: Anaerobicfilterflowschematic Figure 2.9: UASB flow schematic Figure 2.10: tinoieic (CIR:~) acid molecular structure Figure 2.1 1 : Oleic (Cis:r) acid molecular structure Fisure S. 1 2. Steark (C acid molecular structure Fisure 2.13 : Proposed model of fatty acid transport in Escherichia
col; Figure 2.14: Postulated mechanism of the interaction of linoleic (C 1 4
acid with active site of A"-cis, A' l-trans-isomerase and the conversion of substrate to cis-9,trans- 1 1 -octadecadienoic acid
Figure 2.15: P-Osidation cycle showinz individual enzyme reactions Figure 2 16: Two possible pathways of linoleic (cis 9, cis 12))
acid P-oxidation Figure 2.1 7: Two possible pathways of oleic (Cl (cis 9)) acid
(3-osidation Fisure 2.18: Possible biohydrogenation pathway for linoleic (C1s:z)
acid by rumen bacteria Figure 2.19: Proposed a-linolenic (Clp:3) acid hydrogenation mechanism Figure 2.20: Free energy vs hydrogen concentration for some
P-oxidation and methanogenic reactions Percent LCFA estracted into chloroforni Percent LCFA extracted into chloroform:rnethanol Percent LCFA extracted into hexane Percent recovery for 5.625 mg^" C6 to C1g in hexane and hexane MTBE: 5 mins shaking at 200 rpm Percent recovery for 5.625 r n g - ~ - ' CU to CIR in hexane and hexane MTBE: 15 mins shaking at 200 rpm Linoleic (C 1 8 : ~ ) acid extraction partitioning studies Reactors schematic Reactor B glucose byproduct degradation profiles Reactor B gas production profiles Acetate profiles for glucose degradation in serum bottles Linoleic (Cl gT2) acid (LA) degradation profiles
s i i
Figure 5 . 2 :
Figure 5.3:
Figure 5.1:
Figure 5.5:
F ipre 5.6:
Figure 5.7: Figure 5.8:
Figure 5.9:
LCFA concentration profiles for cultures receiving 100 mg^" linoleic (C18:2) acid (LA) Palmitic (Cl6:O) acid profiles in cultures receiving linoleic (C I 8:2) acid (LA) Myristic (Cl.4:o) acid profiles in cultures receiving linoleic (CW) acid (LA) Oleic (CIs:l) acid (OA) profiles in cultures receiving linoleic acid (LA) Effect of linoleic (Cig:2) acid (LA) concentration on acetate production Linoleic acid degradation study mass balance Acetic acid removal in the absence of linoleic (CIEJ:~) acid (LA) Acetic acid removal in the presence of linoleic (c18:i)
acid (LA) Fisure 5.10: Hydrosen removal in the absence of linoleic (Clg:~)
acid (LA) Figure 5.1 1 : Hydrogen removal in the presence of linoleic ( C I P : ~ )
acid (LA) Figure 5.12: Hydrogen rernoval in the presence of linoleic (Cin,~) acid
(LA) 18 days after LA addition F e 5 . 3 : Hydrogen removal in the presence of linoleic ( C ~ M ) acid
Figure 5.14
Figure 6.1 :
Figure 6.2:
Figure 6.3 :
Figure 6.4:
Figure 6.5: Figure 6.6:
Figure 6.7:
Figure 6.8:
Figure 6.9:
Figure 7.1 :
Figure 7.2:
(LA) 35 days after LA addition Proposed pathway for formation of palmitoleic (C16.1) acid from linoleic acid Oleic (CIg:l) acid (OA) profiles in cultures receiving oleic (Cis:~) acid Palmitic (CI6:()) acid profiles in cultures receiving oleic (Cr*:[) acid (OA) Myristic (Ci4:o) acid profiles in cultures receiving oleic (Crx:~) acid (0.4) Effect of oleic (C1g:l) acid (0.4) concentration on acetate production Oleic acid (0.4) degadation study mass balance Acetic acid removal in the absence of oleic (Ci8:l) acid (OA) Acetic acid removal in the presence of oleic (Cix:~) acid (OA) Hydrogen removal in the absence of oleic ( C I K : ~ ) acid (OA) Hydrogen removal in the presence of oleic acid (OA) Stearic (C18:0) acid (SA) degradation profiles in cultures with stearic acid Acetate production profiles for cultures fed with stearic (C 1 RO) acid (SA)
F i g r e 7.3 Figure 7.4
Figure 7.5
Figure 7 .6 :
Figure 7.7:
Figure 8.1 : Fipre 9.1 :
Figure 9.2:
Figure 9.3
Figure 9.3:
Figure 9.6:
Figure 10.1 :
Stearic (CIR:O) acid (SA) degradation mass balance Acetic acid removal in the absence of stearic (Cis:o) acid (SA) Acetic acid removal in the presence of stearic (Cig:o) acid (SA) Hydrogen removal in the absence o f stearic (Clg:~) acid (SA) Hydrogen removal in the presence of stearic (CI 8:0)
acid (SA) Proposed C 18 LCFAs degradation pathways Glucose degradation profiles for cultures with and without diethyl ether Glucose degradation profiles for cultures receiving individual and mixed LCFA substrates Butyrate degradation profiles for cultures with and without diethyl ether Butyrate degradation profiles for cultures receiving individual and mixed LCFA substrates Duplicate control cultures hydrogen profiles with and without diethyl ether Hydrogen profiles for cultures receiving individuai and mixed LCFAs Possible LCFA inhibition pathways
1.0 INTRODUCTION
1.1 Context
Many industrial effluents contain fats and oils in addition to proteins and
carbohydrates. The composition of these effluents is variable and is characteristic of a
particular indusu).. Some effluents contain only fats and oils while in others proteins and
carbohydrates are present. Effluents containing fats and oils originale from edible oil
manufacturing industries (Beccari et al.. 1 996: Harndi er al., 1 992). slaughterhouses
(Sayed el ai.. 1984). fried foods (Landine et al.. 1987). livestock fanns (Broughton er ai..
1998: Hobson er al.. 1981) and dairies (Perle er al.. 1995; Backrnan er al.. 1985).
Treating these effluents containing fats and oils is of concern for aesthetic reasons and
because of the high biochemical osygen demand (BOD) concentration. High BOD
emuents discharged to water bodies pose a threat to aquatic life and to the local
ecosystsm. These effluents can be treated at the site or they can be discharged to
rnunicipaI \\.astavater treatment systems.
Because of the high BOD loadings these wastes may impose on municipal treatment
sq-stems. on-site treatment is desirable to reduce the contaminant level. Biological
treatment. either aerobic or anaerobic. can be used to treat these effluents. Factors such
as BOD loading. foaming and impedance of oxygen transfer affect aerobic ireatment of
long chain fatty acids (LCFAs). However. higher BOD loadings c m be applied to
anaerobic systems and hence. they are more suitable to treat effluents containing fats and
oils. Additionallj.. anaerobic treatment offers advantages of no oxygen addition and
methane recovery.
Anaerobically. fats and oils are biodegraded to LCFAs and glycerol by
microorganisms (Hanaki et al.. 1981). In biological matment systems. LCFAs are
degraded ho~vever. LCFAs are inhibiton. to aceticlastic methanogens (Koster and
Cramer. 1987: Hanaki et al.. 198 1 ). Low degradation rates of LCFAs and the inhibitop
effects of LCFA byproducts as \vell. are major problems facing the development of
suitable anaerobic ueatment technology.
Several rssearchers have conducted LCFA degradation and inhibition studies between
35 to 5 5 OC. Ho~vever. research at louer temperatures is required to better understand the
degradation and inhibition of LCFAs in anaerobic systems. Johns (1 992) reported that
effluents from slaughterhouses in Europe are at approximately 20 OC while in Ausudia.
the? vary between 30 to 35 OC. Between 30 to 55 OC. Hwu (1997) reported that
inhibiton. effects. s lo~v degradation of oleic (Cis 1 ) acid and temperature are important
factors affëcting LCFA degradation. As temperatures were increased from 30 to 5 5 OC.
degradation rates and inhibitoc effects of oleic (Cis 1 ) acid incrmed (Hwu. 1997). For
some industrial effluents containing LCFAs. anaerobic treatment within the thermophilic
range ma? require raising the temperature of the incominp Stream. From an economical
point of vie\\.. adding rnergy ma>- not be a feasible. Hence. for reactors operating at
leu-sr than 30°C. it is important to understand LCFA degradation and inhibition caused
by these acids on anaerobic organisms.
This study. \\-hich focused on the anaerobic degradation of several common LCFAs
(linoleic ( C i s 2 ) . oleic (Cis ,) and stearic (C l s acids) found in fats and oils. therefore.
a a s conducted under room temperature conditions at 21 O C to answer a number of
questions. Would linoleic (Cl* 2). oleic (CI8 ,) and stearic (Cis acids be appropriate
substrates for anaerobic bacteria at 2 1 OC? If so, what are the rates and the pathways of
degradation of these LCFAs? Would these LCFAs and their degradation byproducts
inhibit acidogens. acetogens. aceticlastic methanogens and hydrogenotrophic
methanogens at 2 1 OC?
1.2 Research Objectives
The main objective of this research was to examine the degradation of linoleic (CI8 .).
oleic (Cis !) and stearic (C 18 acids and their inhibitoy effects on acidogens. acetogens
and methanogens at 2 1 i 1 O C .
To meet the main objective. the following specific objectives were developed:
1 . To confirrn the anaerobic degradation pafhrc.a~ and byprodwt distribution for linoleic
(Cls I). oleic (Cls 1) and srearic (Cl8 0) acids.
Degradation of LCFAs proceeds by P-osidation however. it is unclear whether
complete double bond saturation is necessi. for the pathway to proceed. This
objective \\.il1 imtstigate the degradation and byproduct distribution of linoleic
(C 18 .). oleic (C i s 1 ) and stearic (C 18 acids at several substrate concentrations.
2. To determine the inlzibiroc, eflects of linoleic fCls d , oleic (Cls 1) and siearic (Cl* a)
acids. inclitding mixrures of all three acids. on acidogenesis.
Glucose fennenting acidogens are part of the anaerobic microbial consortium and
LCFAs may inhibit glucose fermentation. This objective uill detemine the effects of
Iinoleic (C 18 ?). oleic (C I 8 1 ) and stearic (C s.o) acids and mixtures of al1 three acids at
several concentrations on acidogenesis.
3. Tu determine fhe inhibiroq~ effecrs of linoleic fCls.d. oleic (Cls 1) and srearic (Cls 0)
acids. ird.diilg ntixrures ofal2 rhree acids. on acerogenesis.
The conversion of butyrate. a glucose fermentation byprcduct. to acetate may be
inhibited in the presence of LCFAs. This objective will investigate the effects of
linoleic (Cis,2), oleic (ClS.!) and stearic (Cis:o) acids and mixtures of ail three acids at
sevcral concentrations on acetogenesis.
1. To determine the inhibitor-). effecrs of linoleic (Cls d . oleic (Cl8 1). and stearic (Cl8 O/
acids. and mixtures of ail rhree acids. on aceticlastic methanogenesis.
Acetate deri\-ed from butyrate fermentation and LCFA P-degradation is convened
to carbon dioside and methane by aceticlastic methanogens. However. in the
presence of LCFAs acetate conversion is inhibited. This objective will investigate the
effects of linoleic (Cig ,). oleic (Cls 1 ) and stearic (CIg O) acids and mixtures of al1
thres acids at ss\.eral concentrations on aceticlastic methanogenesis.
5 . To deîerritine the inhibirory effecrs of linoleic (C18.d. oleic (Ci8 and sreuric (Cl8 O)
acids irtclrrdiitg misrzcrrs of al1 rhree acids oit hydrogenorrophic merhanogenesis.
Hydrogen. a b>product of acetogenesis and P-osidation is combined with carbon
dioside to produce methane and water by hydrogenotrophic methanogens. This
objective w-iI1 investigate the effect of concentration dependence of linoleic (Cl8 ?)-
oIeic (Cls i ) and stearic (Cls O) acids and mixtures of al1 three acids on
h)-drogenotrophic rnethanogenesis.
1.3 Thesis Outline
Chapter 1 outlines various problems facing anaerobic wastewater treatment of LCFAs
and based on these issues several objectives are outlined. The Iiterature review presented
in Chapter 2 sections 2.2 and 2.3 briefly summarizes anaerobic reactions and an overview
of several treatrnent technologies is subsequently presented. Section 2.4. the main focus
of the Iiterature review. discusses sources. treatment, composition and structure of
LCFAs. Subsequent sections in Chapter 2 examine LCFA depadation, effects of
hydrogen and volatile farty acids on LCF A degradation. LCFA inhibitory effects and
factors affecting LCFA degradation and inhibition. Chapter 3 provides a description of
al1 materials and methods used in this study. Results for the operation of batch reactors
used to acclimate an anaerobic culture to glucose are discussed iri Chapter 4. Chapters 5.
6 and 7 discuss results for the degradation and methanogenic inhibitory effects of linoleic
(C 1s :). oleic (C 1s ) and stearic (C l 8 acids on methanogenesis. respectively.
Degradation of linoleic (C is .). oleic (Ci i) and stearic (C is acids are compared in
Chapter 8 and a mechanism is postulated. Inhibitop effects of individual and mixtures of
linoleic (Ci8 ?). oIeic (Cis i ) and stearic (Cis acids on acidogenesis. acetogenesis and
methanogenesis are discussed in Chapter 9. Inhibition data from Chapters 5.6, 7 and 9
are compared in Chapter 10 and possible inhibitory pathways are described. S w n m q
and conclusions are presented in Chapter 1 1. FinaiIy, Chapter 12 discusses the
engineering significance of this work and proposes future research that should be
conducted.
1.4 Publications
Several sections of this thesis have been published and subrnitted for publication in
con ference proceedings and refereed journals. Section 5.1.1 of C hapter 5 -0 bas been
publ is hed in Errvrrorrrnerrtal Engirreeririg 1999, Proceedirlgs of the ASCE-CSCE Natiorlal
Coriferetrce on Erw~rorrmenta/ Engiiieeririg (Lalrnan and Bagley, 1 999). In addition,
several sections of Chapter 5 are in press for publication in Water Research. Only
Section 5 . 1 -7.B.ii was escluded from the Warer Research subrnission. Section 6.1. I has
been publ ished in Em*irorlmenta/ E;rlgilleering 1999, Proceedings of the ASCE-CSCE
Il'atiot~al Corfererlce or1 E'wirorlmerttai Engineering (Lalman and Bagley, 1999).
2.0 LITERATURE REVIEW
To farniliarize the reader with anaerobic treaunent fundamentais. a brief introduction
to chemical reactions and treatment technologies is presented in Sections 2.2 and 2.3-
2.1 Overview
.4naerobic treatment is a proven technology and is widely used to treat effluents fiom
man' industries (Spsece. 1996: Wheatley. t 990). The technolog>. has been researched
estensively in order to understand the microbial degradation processes. implernentation
of proper operational strategies and the development of reactor technologies. However.
anaerobic treatment currently faces severaI problems relating to effluents containing fats
and oils. Al though this literature review emphasizes issues arising from the anaerobic
treatment of LCFAs. an oven-iew of anaerobic degradation reactions and a bnef
description of severaI treatment technologies is also discussed.
.A consortium of microorganisrns mediates anaerobic degradation of comples organic
substrates. During degradation. byproducts from one reaction s e n e as substrates for
other reactions in the sequence. Specific microbial populations essential for the process
to function efficiently mediate the reaction sequence. Ultimately. carbon dioside.
methane. \vater and biomass are major end products of anaerobic treatment.
Several reactor technologies currentIy used to treat effluents containing fats and oils
are esamined. A description is provided for selecred reactor configurations including
suspended grou-th. hybrid and attached growth. Finally. this review focuses on LCFA
sources. composition. biodegradation. metabolic byproducts. inhibitory effects and
factors affecting degradation and inhibition.
2.2 Fundamentals of Anaerobic Wastewater Treatment
2.2.1 Hydrolysis
Conversion of carbohydrates. proteins and lipids anaerobically to methane is linked by
several reactions. each mediated by a specific microbial population. The process, sho~-n
in Figure 2.1. is divided into hydrolysis. acidogenesis. acetogenesis and rnethanogenesis.
Hydroiysis. the first reaction step. is the degradation of complex organic pollmers into
monomers. Conversion of a polysaccharide (Figure 2.2) into glucose by hydrolsic
bacteria is catalyzed by arnalyse. an extra-cellular enzyme (Lehninger er al., 1999).
Cellulose- protein and lipid hydrol>-tic reactions are cataiyzed by cellulases. proteases and
lipases. respsctivel>-. Several esarnples of hydrolytic reactions and their negative free
energies reported in Table 2.1 show that organic polymers are hydrolyzed to monomeric
compounds.
Factors affecting enzymatic hydrolysis rates include substrate solubility. temperature.
pH. and the t>.pe of substrate (McInerney. 1988: Gujer and Zehnder. 1983). In contrast
to proteins. hemicellulose and lipids have higher hydrolysis rates. For example.
hydrol>-sis rate constants for hemicellulose. lipids and proteins are 0.54 (d-') (Ghosh et
al.. 1980). 0.4 - 0.6 (d-') (Heukelekian and Mueller. 1958). and 0.02 (d-') (Woods and
.Melina, 1965). respecti\.ely. Hemicelluiose hydrolysis occurs when cu(l-+ 6) o rp ( l+ 4)
ether linkages are esposed to hemicellulases.
Lipids and proteins uith hydrophobic components are less hydrophilic than
carboh>.drates and do not dissolve rsadily into the aqueous phase. During hydrolysis.
lipids and proteins are cleaved at ester and amide bonds to produce LCFAs and amino
acids byproducts. respectively (Figures 2.3 and 2.4). Protein hydrolysis rates are less
than hemicellulose because of the comples folding structure in complex proteins
(Lehninger er al.. 1999; Ghosh et al.. 1980: Woods and Melina. 1965). Complete
hydrolysis of peptide bonds is accomplished only after the quatemary or tertiary protein
structure unfolds to expose the primary amide bonding structure.
+ Lactate. propionate.
butyrate. ethanol, etc. -
-
Carbohydrates. proteins and lipids
v
Figure 2. I : .c\naerobic conversion processes (Gujer and Zehnder. 1983)
Long Chain Fatty Acids
(LCFAs)
Sugars
-
v Acetate
1 A
* CH& CO?
Amino Acids 5 -
-4 simple tripeptide protein \\ith its p r ï m q amide bonding structure is shown in
Figure 2.3. Under unfavorabie conditions' unfoidine the protein is not
Glucose monomer
O a ( 1 - 6 ) linkage
/
Figure 2.2: Structure of a polysaccharide (Lehninger ei al.. 1999)
Table 2.1 : Some hydrolytic reactions and free energies AG, (w-mole-' )
Sucrose +- H 2 0 - D-fnictose + 0-D-glucose -43 -6 ' GI>qfglycine -+ H 2 0 - 2 Glycine -9.2 ' Frtx rnergy values taken from Thauer et ai. (1977)' and Lehninger er al. (1 999)' Al1 free energ)- values reported in this table and other sections of this test are based on standard conditions at 25 OC and unit concentrations. AG,' is the reaction free energy adjusted to pH 7.
thermodynamically feasible because hydrophilic molecuIes on the outer protein structure
hide hydrophobic amino acid components of the structure within the complex tertiary and
quaternary cornplex (Lehninger et al., 1999). The rate determining step for protein
degradation is hydrolysis of the structure into free arnino acids (Heukelekian, 1958).
Lipids are hydrolyzed into LCFAs and glycerol by esterase enzymes. An example of
a typicaI lipid structure is shown in Figure 2.4. O'Rourke (1968) reported that lipid
hydrolysis was rate limiting during sludge digestion within the pH range fiom 6.7 to 7.4.
Carbosyi- terminal end 0
Alanine v
! O O CH3
I
bond amino acid
. CH-N I CH20H
Serine
residue
Amino- terminal end
Phenylalanine
Figure 2.3: Structure of a tripeptide (Lehninger et al.. 1999)
Figure 2.4: Structure of a triacylglycerol (Lehninger et al.. 1999)
Eastman and Ferguson (1 98 1 ) concluded that lipids are not degraded during sludge
digestion at pH 5.2. Using the first order rate constant as a rneasure of lipid hydrolysis.
OIRourke ( 1 968) reported increasing hydrolytic rates as the temperature increased.
The optimum pH for hydrolysis is variable and for carbohydrate degradation to
ducose. the masimum hydrolysis rate occurs at pH from 5.5 to 6.5 (Zoetemeyer er al.. C
1982). For proteins. the optimum pH is 7 and higher (Breure and van Andel. 1984) and
for lipids. the optimum pH has not been reported.
Several researchers have used kinetic models to fit data from hydrohsis studies. 3-oike
er al. (1 985) investigated fitting anaerobic degradation profiles for cellulose and starch to
kinetic models. Noike er al. (1 985) used the Contois-Chen mode1 to fit cellulose
degradation and the Monod equation to mode1 hydrolysis of starch into glucose. Based
on results from these models. cellulose hydrolysis was reported as the rate-limiting step.
The cause of the lo\v reaction rates may be related to the inability of cellulases to
hydrolyze cellulose into monomer units because of low cellulose solubility in solution.
2.2.2 Acidogenesis
Formation of acetate. propionate. n- and iso-butyrate. lactate. valerate and ethanol.
from carbohydrates. amino acids and LCFAs is the next step in the reaction sequence
follou.ing hydrolysis (Sahrn. 1983). Examples of several acidogenic reactions are shown
in Table 2.2. Hydrogen production accompanies the formation of acetate. propionate and
but>.rate from glucose. When hydrogen accumulates. formation of reduced VFAs
predominates to maintain low hydrogen partial pressures (pHz). Hydrogen removal by
sulfate reducing hydrogenotrophs or hydrogenotrophic methanogens influences
Table 2.2: Some acidogenic reactions and fiee enernies AG, (k~-mole-')
have been used to treat effluents from slaughterhouses (Dague et a(.. 1990). potato
processing (Landine et al.. 1987): dairies (Landine et al., 1988) and edible oils
(Southu-orth. 1979). Operating data for full-scale low rate lagoons provided in Table 2.5
show that these systems are loaded between 0.1 to 2 kg COD^"-^-'.
Table 2.5: Design parameters reported for anaerobic reactors treating eMuents containing fats and oi!s
Reactor Waste HigMow Scale Temp. COD CH4 Ref. Type rate m3 O C load - vield
kpem4.d-l
Lagoon Potato processing low 45000 27 0.13 0.3" 1 Lagoon Edible oil low 68000 45 0.6 - 0.9 - - 7 UASB (floc) Slaughter house high 10 30 5 3a 3 UASB S laughter house high 33 20 7 2.8" 4 (granulated) U.4SB Dairy high 400 30 7 - 5 A F S laughter house high 83 36 8 0.32~ 6 AC S lauphter house hi& 11120 35 3 0.24' 7
kg C H ~ - C O D - ~ ~ - ~ - ~ - ' ; kg CI& +kg COD" removed; m' biogas -kg COD" added; AF = anaerobic filter AC = anaerobic contact 1 Landine et al. (1 987); 2~outh\vorth (1 979); %ayed and de Zeeuw (1 988); 'saYed er al. (1 987): '~amson er al (1985); 6 ~ e t z n e r and Temper (1990); 'stebor er cd. (1 990)
2.3.3 High Rate Treatment
High rate systems treating effluents containing LCFAs include upflow anaerobic
sludge blanket (UASB) (suspended growth) (Sayed and Zeeuw, 1988), U.4SB
(granulated sludge) (Hwu. 1 997; Sayed et al., 1 987), anaerobic filter (Metzner and
Temper. 1990). anaerobic contact (Stebor er al.. 1990) and anaerobic sequencing batch
reactor (anSBR) (Dague and Pidaparti. 1992). Figures 2.7 to 2.9 show several reactor
Solid 1 Liquid Separator
Flocculator influent
Anaerobic reactor
7- Effluent
1
Sludge Recycle
Figure 2.7: The Anaerobic contact process flow schematic
Biogas
Biomass attached support material
J
Recycle
Figure 2.8: Down flow fixed film anaerobic filter schematic
configurations used to treat effluents containing LCFAs. Operational data for bench and
full scaIe installations are provided in Table 2.5.
High rate reactor configurations offer several advantages over the low rate lagoon
system. For exarnple. in the contact process (Figure 2.7), a flocculator is added following
the anaerobic reactor to assist in soiid / liquid separation (Schroepfer and Ziemke. 1959).
This reactor design configuration ailows for longer solid retention tirnes (SRTs).
-4naerobic sequencing batch reactors (anSBRs) (Dague, 1992) offer the advantage of
operational flesibility over continuous treatrnent systems. Because the anSBR reactor is
operated in batch mode. filling, reacting, settling and decanting are accomplished in a
sinzle vessel. The simplicity in design allows for no separate clarifier and no extemal
biomass \vashout through immobilization and a quiescent inlet region where large dense
biofilms are developed (Speece. 1983). The downflow configuration is normally used to
treat effluents containing linle or no suspended solids while operation in an upflow mode
is used to prevent plugging caused by emuents containing high suspended solids.
Howe~~er. lab scale studies have shown effluents containing high suspended solids (1 3
CL-') can be treated in the downflow configuration (Kennedy and Guiot. 1988). Another b
ds\vloped technolog!.. which involves no support for biomass immobilization. is UASB
reactors (Figure 2.9) (Lettinga and Hulshoff, 1992; Lettinga et al.. 1980). Dense granules
in UASBs avoid the added cost of packing material, which is necessary for biomass
retention. GranuIes with their high settling velocities and good settleability are
distinctive features of UASB reactors. Under some circurnstances, an additional
gas :' solid separator
I
Effluent
Recycle
Influent A 7
Figure 2.9: UASB flow schematic
clarification process is added ont0 the emuent to aid in soIid/liquid separation. Typical
loading rates for high rate reactors vary between 5 to 25 kg COD-^"-^-' (Hall. 1992).
2.4. Long Chain Fatty Acids
2.4.1 Sources and Treatment
Fats and oiIs are present in effluents from daines (Perle et al.. 1995)- slaughterhouses
(Sayed and Zeeuu.: 1988. Sayed et al.. 1987; Sayed et al.. 1984), [ivestock farms
(Broughton er al.. 1998; Hobson et al.. 198 l), and edible oil processing facilities (Becker
er al.. 1999: Beccari er al.. 1996; Hamdi et al.. 1992). As previously discussed. the
presence of these compounds in industrial effluents causes several problems for aerobic
treatment systems. The high COD in these emuents exceeds the allowable organic
loading to aerobic systems. In addition, LCFAs are surfactants and as such cause foam
and scum formation (Lemmer and Baumann. 1988). They also f o m an oily film around
microbial flocs. impeding osygen transfer and decreasing the efficiency of aerobic
organisms (Becker er al.. 1999).
Sekreral difficulties also arise during the treatment of fats and oils in high and low rate
anaerobic reactors. Stebor et al. (1 990) reported that effluents containine fats and oils
caused anaerobic reactors to operate inefficiently because of problems related to low
solubility. Io\- degradation rates and solids (fats and oils) flotation. Studies by Hnu
( 1997) reported sludge flotation during treatment of oleic (Cis !) acid in a UASB reactor.
In addition. fats and oils are inhibitory to anaerobic organisms (Broughton et al.. 1998:
Koster and Cramer. 1987. Hanaki er al.. 198 1 : Novak and Carlson. IWO).
During treatment. the first reactior, step is hydrolysis of fats and oils into free glycerol
and LCF.4s by lipases (Hanaki er al.. 1981). Under anaerobic conditions. glyceroI is
degraded to 1.3 propanedioi (Biebl et al., 1998) and subsequently to acetate (Qatibi el al.,
199 1 ). Hou.s\.er. LCFAs are not easily degradable and are inhibiton. to aceticlastic
methanogens (Rinzema et al.. 1994: Angelidaki and Ahring. 1992: Koster and Cramer.
1987: Hanaki et al., 198 1 ). Additionally. Angelidaki and Ahring (1 992) reported LCFAs
are inhibiton. to organisms consuming propionate and butyrate.
2.4.2 Composition and Structure
PhysicaI properties and chemical composition for several LCFAs used in this stud!.
are described in this section. Fat and oils are glycerol esters of fatty acids and the
predominant ester is the triglyceride which consists of a glycerol backbone with three
long chain fatty acids (Fonno. 1979). The molecular structure of a typical
triacylglyceride was previously shown in Figure 2.4. The less common di- and
monoglycerides consist of two fatty acids and one fatty acid molecule. respectively. I t is
difficult to distinguish between glycerides of fats and oils and according to Forrno
"Reversible changes in the state owing to variation in temperature may obliterate the common conception that fats are solids and oils are liquids. so today this distinction between the terms fat and oil is largely academic. The tems are still used comrnercially, but the? have on1 y limited significance".
Physical properties of linoleic (C 1 8:2). oleic (C 8: 1) and stearic (C 1 S:O) acids. three
LCF.4s used in this study, are shown in Table 2.6. Further data on physical and chernical
characteristics of other LCFAs is available in "Baileys' Industrial Oil and Far Producrs"
by A.E. Baile'. (1979). At room temperature. both linoleic (C18.z) and oleic (C18 !) acids
are liquid u-hile stearic (Cis acid is solid.
Tabie 2.6: Selected properties of linoleic (C18.1), oleic (C18:1) and stearic (CI8.0) acids LCFA Melting Point ( O C ) Boiling Point (OC) Densiîy
Biodegradation of LCFAs proceeds via several steps including adsorption ont0 the
ce11 surface. movement across the ce11 surface and LCFA conversion to lower molecular
~veight components.
Adsorption onco the ce11 surface
The first step of the LCFA degradation process begins with adsorption ont0 the ce11
membrane. Adsorption is a physical or chernical process controlled by factors such as
temperature. pH. rnultivalent cations. agitation speed. agitation time and adsorbate
polarity (M>.ers. 1999: Daniels. 1980). The effect of temperature on adsorption is
variable and depends on the nature of the adsorbate and adsorbent. For example. as
temperature increases. the "stickiness" of surfaces decreases causing adsorbed ionic
surfactants to desorb (Myers. 1999). This type of adsorption can be modeled as a
physical process b>- considering surface forces to become weaker due to increased kinetic
energies as surface temperature increases (Adamson and Gast. 1997). In cornparison. in
some systems containincg nonionic surfactants. adsorption increases as temperature
increases (Tilberg and Malrnsten. 1993). The beha\ior of nonionic surfactant adsorption
is c o n t r e to ionic surfactants and can be modeled as a chemical process. As
temperature increases to an optimum value. chemical bonding between nonionic
surfactants and a surface increases and beyond an optimum temperature. desorption
begins to occur (Somasundaran er al., 1997).
pH strong1~- influences the adsorption of bacterial cells ont0 soils. clays. hydrous
metal osides and ion eschange resins (Daniels. 1980). Altering the pH reverses
adsorption once the adsorbate is attached to a surface. Adsorption of proteins ont0
surfaces is influenced by changes in pH. which affects the isoelectric point and modifies
the surface charge. Generally. stronger adsorption occurs between pH 3 to 6 (Daniels.
1980). The effect of pH on LCFAs adsorption ont0 microbial cultures has not been
in~eestigated.
Adsorption is affected b>- the binding of cations ont0 adsorbate or adsorbent.
Multi\.alent cations compete with adsorbates for active sites in solution and the addition
of inorganic salts promotes desorption (Daniels. 1980). Several studies have esamined
addition of CaC17 to decrease LCF.4 inhibition (Koster. 1989; Hanaki et al.. 198 1 ;
Galbraith er al.. 197 1). The role of ~ a - ' ions is unclear but binding of ~ a - ' ions with
LCFA carboxylic g o u p s is likely the mechanism preventing adsorption and hence.
inhibition.
.4n increase in agitation speed and reaction time enhances adsorption by increasing the
probability of surface contact. Generaliy, equiIibrium is reached within 15 minutes.
although in some cases, it may take a fe\v hours (Daniels, 1980). Afier equilibrium. if a
system is agitated violentl>. for long penods. the adsorbate may begin to desorb (Daniels.
1980).
Because LCFAs possess hydrophobic and hydrophilic components. their surface
orientation may be different. For esample. if the ionic component dominates at the
surface. the interaction wiil be such that the hydrophobic component is oriented towards
the solution. Han-ever. if the hydrophobic component prevails at the surface. the
h>.drophilic component is directed towards the solution (Ulrich and Stumrn. 1988). On
non-polar surfaces. adsorption of LCFAs is dominated by hydrophobic properties of the
nioleculs and the estent of the interaction increases with increasing hydrophobicity of the
carbon chain (Wrich and Stumm. 1988). Ho~vever. on polar surfaces. the ionic end of the
LCFA binds to the polar surface and the hydrophobic component is oriented towards the
solution.
Surfaces are classified as either as homogenous or heterogenous. Biological surfaces
are heterogenous in nature and made up of comples structures of carbohydrates. proteins
and lipids (Shinitzky. 1993). Adsorption of LCFAs ont0 these surfaces is complicated
and researchers have attempted to use homogenous surfaces to characterize the process.
On hornogenous surfaces such as rke hull, it is reported that addition of an isopropanol
cosolvent influenced the adsorption of LCFA emulsions (Proctor and Palaniappan. 1990).
Procror and Palaniappan (1 990) proposed that hydrogen bonding of the isopropanol
cosolvent to weak acidic sites on the adsorbent promoted LCFA binding by hydrophobic
interaction.
In contrast to rice hull surfaces. biological surfaces have variable adsorption sites and
characterization of the adsorption process is comples. Using aerobic and anaerobic
sludges as adsorbents. Hrudey (1 982) and Hanaki et al. ( 1 98 1 ) reported that LCFAs were
removed frorn the aqueous phase within 20 minutes and 24 hours. respectively. More
recent studies at pH 7.2 and 35 O C have shown that the adsorption of LCFAs ont0
eranulated anaerobic sludge follo\vs the Freundlich isotherrn model (r = 0.992) (Hwu. k
1997). H m (1 997) characterized the process and proposed a biosorption model
consisting of the follouing stages: 1. Adsorption. 2. Desorption. 3. Desorption and
degradation. and 4. Degradation.
Mo~wement across the ce11 surface
.4t the esterna1 surface of the outer membrane. LCFAs transverse the ce11 into the
periplasmic space via a trammembrane membrane protein (NuM, 1986). This
mechanism has been researched extensively by biochemists and ce11 physiologists
because LCFAs are important substrates for mammalian ceIl groWh (Mangroo et al..
1995). LCFAs are used for energy storage (triclyceride synthesis in mammalian cellsj.
lipid synthesis (in prokaryotic and eukaryotic cells) and energy production viap-
osidation (in prokaryotic and eukaryotic cells). The bacterium Escherichia coli has been
snidied as a model to investigate the LCFA uptake mechanism across ce11 membranes
(Mangoo er al., 1995).
Four steps have been identified in LCFA uptake by E. coli (Mangroo et al.. 1995). A
schematic membrane transport mechanism is showm in Figure 2.1 3. FadL, a transversal
outer membrane protein mediates LCFA movement across the membrane. Initiaily.
ionized LCFAs bind to FadL and are then transported across the peptidoglycan layer into
the periplasmic space. The transport mechanisrn across the peptidoglycan layer has not
been elucidated but it is possible a protein may be involved in mediating LCFA transfer
across the layer.
f -Medium-chain fatty
acids (C7-C 1 1)
FadL
acids (C 12-C 18)
OM
ACS Long chain fatîy acid ac y 1-CoA
COA' + ATP
cytoplasm
Figure 2. 13 : Proposed model of fany acid transport in Escherichia coli C7 10 C 1 fatty acids traverse the outer membrane via a membrane protein (FadL). C7 to CI I fany acids also enter the cell by diffusion. Fatty acid becomes activated by acyl-CoA synthetase (ACS) protein to f o m long chah fatty acid-CoA. OM = outer membrane; PG = peptidoglycan; PS = periplasmic space; IM = inner menbrane; FadL = membrane protein: ACS = acyl synthease Co.4. (Adapted fiom NUM, 1986)
-4 protein identified as Tsp has been proposed to facilitate binding and releasing of
LCF.4s across the periplasmic space (.4zizan and Black. 1994). Afier passing througb
the peptidoglycan layer. LCF.4s are protonated in the periplasmic space and diffuse to the
i ~ e r membrane. On the outer part of the inner membrane adjacent to the periplasmic
space. another proposed transversal inner membrane protein anchors protonated LCFA
molecules for activation by adenine triphosphate (ATP). Finally. on the inside of the
imer membrane. acyl-CoA synthetase activates fiee LCFAs into long chain acyl-CoA
cornpieses using ATP.
LCFA Siodegradation
Hydrogenation and osidation are two reaction stages identified during LCFA
degradation. It is unclear whether complete LCFA doubie bond saturation is necessary
before P-osidation. No\& and Carlson (1 970) postulated cornplete LCFA saturation is
required before P-osidation. Hom-ever. Canovas-Diaz et al. (1 99 1 ) reported unsaturated
saturation is the only pathway. then stearic (Crs:o) acid may enter into the P-oxidation
reaction. Ho\vever. in cornparison. for unsaturated LCF.4s. hydrogenation before P-
osidation will be required assuming the postulate b>- Novak and Carlson ( 1 970) is
correct.
Fujimoto et al. (1 993) reported linoleic (Ci8 ?) acid is first hydrogenated to form
several intermediate compounds. Several organisms listed in Table 2.1 1 are reported to
mediate hydrogenation of linoleic (Ci 8:z) acid. Fujirnoto et al. (1 993) also proposed
several possible pathways leading to the formation of stearic (Ci8:0) acid via the
formation of tram- 1 1 -0ctadecenoic acid. A mechanisrn postulated bp Harfoot (1 9 7 8 ) for
the formation of the trans intermediate is shown in Figure 2.14. In the active site. the
substrate is anchored to hydrogen bonds and n -electron interaction sites on the isornerase
enzyme. A series of proton transfer reactions is initiated by hydrogen bond formation at
the carbosyl end of the substrate molecule. Eventually. the conjugated diene product is
rt-leased ~vith the T-bond at carbon number 12 shifiing to carbon number 1 1.
Table 2.1 1 : Hydrogenation of linoleic (Cis -) acid b'v rumen bacteria (Fujimoto er al.. 1993)
Bacterium H ydrogenation Genus and species identification croup products
1 trans- 1 1 -C 18 1 Blrryrivibrio jibr isolvens I I trans- 1 1 -C 1 1 and Cls.o Unidentified II 1 trans-9 C i 8 1 B~cryri\:ibriojibrisolvens
Selenomonas ruminanhm
Se\.eral researchers (Kemp er al.. 1975: Rosenfeld and Tove. 1971 : Viviani. 1970)
have attempted to idsntifi hydrogen donors responsible for hydrogenation of unsaturated
LCF.4s. Viviani (1 970) tested several hydrogen donors and only pyruvate and formate
had a positi\-e effect on hydrogenation acti\.ity. Sirnilar experiments by Kemp er al.
( 1975) failed to demonstrate any effect of pyruvate. formate. succinate and a-
ketoglutarate on hydrogenation. In addition' using labeled substrates. Rosenfeld and
Tove ( 1 97 1 ) reported tritium was not the source of hydrogen for the formation of trans- 1 1
octadecenoic acid from linoleic (C is ?) acid.
The hydrogenation mechanism from oleic (Cl* ,) acid to stearic (Cis acid is not clear
and is based on the observation of the trans-1 1-octadecenoic acid byproduct. Although
Hydrogen bonding site
H-O
Enq-me site for interaction with x-electrons
H>.drogen donor site
Figure 2.14: Postulated mechanism of the interaction of linoleic (C 1s .) acid with active site of A '"cis. A ' ' -trans-isornerase and the conversion of substrate to cis-9.trans- 1 1 - octadecadienoic acid. (Harfoot. 1978; Kepler er al.. 197 1 )
no intermediates have been identified for oleic (Cis 1 ) acid hydrogenation. Kemp er al.
( 1975) suggested rhat the formation of 1 O-hydrox). stearic (Ci* acid may be a key
intermediate. More recently. several researchers ha\*e reponed the formation of 10-
hydrosy stearic (C acid from linoleic (C is ?) acid (Koritala and Bagby. 1992). It is
possible that hydrogenation or hydro1)-sis of unsaturated LCFAs may be possible routes
for detosification prior to metabolic osidation.
LCFA oxidation proceeds microbially via several reactions shown in Table 2.12
(Ratledge. 1991: Mackie er al.. 1991). Al1 five oxidative pathways have been observed in
pr0ka.q-otic and e u k q o t i c orpanisms but the predominant pathway in anaerobic cultures
is P-osidation. The first four pathways in Table 2.12 are discussed briefly and a detailed
analysis is presented for the P-oxidation reaction scheme since this is the major pathway
of concem in this study. In the first pathway, 2-methyl ketones are formed via terminal
carbos>.lic groups \\.hich are reduced with subsequent oxidation of the a-carbon ( i x . the
carbon in the &-position relative to the original carboxylic group). In eukaryotes such as
yeasts. carbon dioxide is produced as the terminal carboxylic group is cleaved during cu-
osidation. During w-osidation LCFAs are converted fiom mono-carboxylic acids to
hydres). carboxylic acids and in (w- 1 ) and in (w-2) oxidation- hydrosy fatty acids are
also fonned. Mid-chain oxidation of LCFAs such as oleic (Cis:i) acid leads to the
formation of 1 O-hydroxy stearic acid (Koritala and Bagby. 1992).
After complete or incomplete biohydrogenation. LCFAs enter into the P-osidation
reaction scheme. For e v e c P-oxidation reaction cycle sho\\n in Figure 2.15. N o carbons
Table 2.13: LCFA oxidation reactions Formation of 2-methvl ketones
Roy et al. ( 1 986) and Kemp and Lander ( 1 984) have investigated degradation
pathways for unsaturated LCFAs. For linoleic (C 18.z cis-9,cis- 12) acid two degradation
pathways (Figure 2.16) are possible (Roy et al.. 1986). One pathway (the left one in
Figure 2.16) starts with three P-osidation cycles to f o m cisxis-3.9 dodecadienoic (C 2
Linoleic acid C 18:2 (cis-9. cis- 12)
Figure 2.16: Two possible pathways of linoleic (cis 9, cis I I ) ) acid P-oxidation. Enzymes: 1. A'-cis-A'-trans enoyl Co A isomerase: 2. A'-cis enoyl Co A hydratase: 3. 3-hydroxyacyl CoA epimerase; 4. A"-c~s-A' '-uans isomerase. (.4dapted from Roy ef al.. 1986)
(cis-3 . cis-9)) acid. Further 9-oxidation is prevented because the structure shoun in
Figure 2.1 6 is unable to undergo dehydrogenation by fatty acyi-CoA dehydrogenase.
This dehydrogenase enzyme is specific to removing two hydrogen atoms from the a and
carbons with subsequent formation of the trans isomer. Hydrogenation by a
hydrogenase or isomerizations by an isomerase are the two choices available for the
reaction to proceed. Isomerization takes place with the formation of trans.cis-2.6
dodecadienoic (C acid which enters the P-osidation cycle tsvice eventually forming
cis-2 octenoic (Cs 1 ) acid. Eventuall~.. cis-2 octenoic (CS 1 ) acid is hydrogenated and the
unsaturated octanoic (Cs:o) acid molecule is hydrolyzed to 3-D-hydroxy octanoic (Cs)
acid. An epirrierïzation reaction transforms 3-D-hydroxy octanoic (Cg) acid into 3-L-
hydrosy octanoic (Cs) acid which then undergoes P-osidation.
LCF-4 P-osidation proceeds via a sequence of reactions to acetate and hydrogen
(Reaction 2.2). In the first step of the sequence. the P-ozridation of stearic (Crg O ) acid to
palmitic (Cl6 acid is show% in Equation 2.4.
Equation 2.3 is controlled thermodynarnically by the hydrogen partial pressure and
assuming 1 M concentration for the acids and a pH of 7. the hydrogen dependence is
derived from equation 2.5 (where AG" is the reaction free energy under standard
conditions and adjusted to pH 7, R is the gas constant. T is the temperature and Q is the
reaction quotient). The final espression is shown as Equation 2.6. Some fiee energy
expressions for several reactions considered for this study are shown in Table 2.17 and
plotteà in Figure 2.20.
AG' = AGO' - RT ln Q (2.5)
AG' = 50.82 + I 1 .l log [HZ] (LJ-mole-') (2.6)
The conversion of stearic (C18.0) acid to palmitic (C16-0) acid is possible over a wide
range of hydrogen partial pressure and the reaction becornes more thermodynamically
favorable as the hydrogen partial pressure decreases. In comparison, the conversion of
acetic acid to methane and carbon dioxide is independent of the hydrogen partial
pressure.
The reaction free energies s h o w in Table 2.17 suggest that P-oxidation of linoleic
(Cl8.?) and oleic (Ci8-*) acids into palmitic (Ci6.0) acid is themodynamically more
fa\*orable in comparison to stearic (Cl* acid. a saturated C 18 LCFA. The reason being
that hydrogen produced during P-oxidation is possibly used to hydrogenate the carbon
double bond hence. lowering the hydrogen partial pressure and favoring the formation of
palmitic (C i6 acid.
2.5 Summary
The development of suitable anaerobic treatment processes to remove LCFAs from
industrial effluents has progressed to a stage where issues relating to the degradation
mechanism and inhibiton. effects remain unresoived. A complete understanding of the
LCFA degradation reaction mechanism has still not been clearly established. Novak and
Carlson ( 1 970) proposed only saturated LCFAs initiate the P-oxidation degradation
mechanism. however. further studies by Canovas-Diaz et al. (1 99 1) have shown oleic
(C 18 1 ) acid may also enter into the reaction scheme. Hence, it is unclear whether or not
cornplete LCFA saturation is required to initiate the P-oxidation reaction. The impact of
LCFAs on anaerobic organisms has been investigated but the reported research has
focused mainly on the inhibition of aceticlastic rnethanogenesis. Additional work
Figure 2.20: Free energy vs hydrogen concentration for some P- osidation and rnethanogenic reactions. (Al1 reactions are numbered in accordance with Table 2.17. Curves for reactions 3 and 5 are approsimately the same)
is ho\ve\.er. required to identiQ the effects of individual and mixtures of LCFAs not only
on aceticlastic mcthanogens but also on acidogens. acetogens and h>.drogenotrohphic
methanogens.
Degradation and inhibition studies have been mainly focused at temperatures ranging
from 35 to 55 O C . In contrast. effluents at 20 OC or less will require additional energy to
reach a desirable reactor target temperature within the mesophilic or thermophilic range.
In some cases. increasing the reactor temperature within the mesophilic or thennophilic
range is not possible because of process or economic reasons, Therefore, developing
anaerobic technologies operating at lower than mesophilic temperature is required to treat
effluents containing LCFAs.
3.0 in4TERIALS AND METHODS
3.1 Experimental Pian
The esperimental pian was developed in accordance with the objectives of this
research. An experimental design matrix outlining the experiments conducted during this
work is shown in Table 3.1. Degradation experiments shown in Table 3.2 were designed
to confinn the LCF.4 P-oxidation mechanism and at the same time to determine LCFA
inhibition of the biomass. The effects of LCFA concentration on the f3-oxidation
mechanism were exarnined during the degradation studies.
Inhibition studies (acidogenesis, acetogenesis, aceticlastic methanogenesis and
hydrogenotrophic methanogenesis) were designed to investigate the effects of LCFA
concentration on the degradation of substrates specific to each population. Experimental
details for the three C l g LCFA studies are s h o w in Tables 3.3 to 3.7.
Note: O denotes experiment not perfonned, 1 denotes experirnent perfonned
Table 3.1 : Esperimentd design matrix
Details of interaction esperiments to determine the additive effects of individual and
mixtures containing Iinoleic acid, oleic (Cr8:I) acid and stearic (Cis:~) acids on
acidogens. acetogens and hydrogenotrophic methanogens are shown in Table 3.7. Al1
controls tvere prepared in duplicate and sarnples containing LCFAs were prepared in
triplicate. The culture used during the duration of this study was acclimated to glucose
(99.999%) and hydrogen (99.999%) gases (BOC Gases, Toronto, ON) were used to
calibrate the gas chromatograph (GC). Carrier gases used were helium (99.999%) and
nitrogen (99.999%). Glucose (BDH Chemicals. Toronto. ON) degradation was
monitored with a glucose hesokinase kit (Sigma Chernical Co., St. Louis, MO). The kit
consisted of 0.75 mrnole NAD. 0.5 mmole ATP, 500 units glucose hexokinase (yeast),
500 units glucose-6-phosphate dehydrogenase. 1.05 mmole magnesium ions and 0.75 g
sodium azide.
3.3 Batcb Reactors
3-3.1 Inoculum Source
Seed culture to MO batch reactors was prepared using a 1 :6 mixture of anaerobic
digester sludge and granulated anaerobic biomass collected from the Toronto Main
Treatrnent Plant and a food processing plant in Cornwall. Ontario. respectively. A 4-L
somi-continuous reactor (Reactor A) mith a 3-L liquid volume was maintained at 2 1 OC
with 20.000 r n g - ~ - ' volatile suspended solids (VSS). Using basal medium at pH ranging
Table 3.8: Basal medium cl Parameter
K2HPOj (N&)~SOJ NaHCO; NHjHCO; h4gC12.4H20 KCl HjBOj FeC12.3H20 ZnCl? MnC 12.4H20 C U C ~ ~ . ~ H ~ O ( N H ~ ) ~ M O O ~ . ~ H ~ O C O C ~ ~ . ~ H ~ O NiC12.6Hz0 NazSeO; EDTA Resazurin Yeast extract
(adapted from Weigant and 1
racteristics Concentration. rnp L"
from 8.0 to 8.2 (Table 3.8), inoculum fiom Reactor A was diluted to 1500 r n g ~ - ' VSS
into a second 3-L semi-continuous reactor (Reactor B). Reactor B was also maintained
with a 3-L liquid volume. Biomass from Reactor A served as an inoculum source to
reactor B as needed and inocula for the serum bottles (1 60 mL) were collected from
Reactor B.
3 - 3 2 Operation of Inoculum Reactors
Rcactors A and B were operated in batch mode and fed with 1000 rng -~ - ' glucose
every 5 to 6 days (time when acetate and gas production measurements indicated that al1
glucose and byproducts were consumed). Glucose feed solution (30.000 mg^-') for both
reactors A and B \vas prepared in basal media. Operational stability for both reactors \vas
monitored using pH. alkalinity (as CaCO;) and VF.4 measurements. Prior to inoculation
of the 160 mL senun bonles. VFA concentrations. pH. alkaIinity (as CaCO;). total
suspended solids (TSS) and volatile suspended solids (VSS) were measured to
characterize the biomass in Reactor B. VFAs were measured to ensure no residual
remained in the reactor. The semi-batch reactor (Reactor B) was operated with a HRT of
approsimately 150 days and an organic loading rate of 0.21 g COD-L"-d-'. In
cornparison. Maillachemvu and Parkin (1996) reported a loading rate of 0.5
g COD-~ ' ' -d - ' to a culture in a batch reactor.
3.4 Hydrogen and Methane Measurement
Head space gis samples (20 PL) for hydrogen and methane were analyzed using a
He~vlett Packard 5890 gas chromatograph equipped with a thermal conductivity detector
(TCD) and a 30-m x 0.53-mm diameter CarboxenTM (Supelco) plot column. Analysis
u-as isothermal at 60 OC with nitrogen as the carrier gas at 5 ml-min" and the detector
and injector temperatures set at 250 OC and 200 O C , respectively. Hydrogen and methane
were detected at 1.19 and 1.78 minutes. respectively. The detection limit for hydrogen
\vas 0.063 3 kPa and for methane it was 0.484 kPa,
Calibration standards for the gas chromatograph were prepared in senun botrles (160
mL) that had been purged with nitrogen (99.998%) for 2 to 3 minutes. The bonles were
sealsd uith ~eflon' lined septa and capped with aluminum crimp seals. Knowm
quantities of hydrogen and methane were injected into the capped bottles. Triplicate
sarnples (20 PL) were prepared for each gas concentration measured. Calibration curves
for hydrogen and methane were prepared using gas samples ranging from 0.0633 to 3.17
kPa and 0.0483 to2.40 kPa. respectively.
Prior to analyzing headspace samples. a blank and two standards were prepared within
the calibration rangs and ana1)zed for carbon dioxide and methane. Gas standards were
also anal>zed after every 10 samples to ensure the instrument remain calibrated. During
the duration of the study, analyses of the standards were found to lie within less than 5%
of the calibration cunTe.
3.5 Volatile Fatty Acid (VFA) Measurement
During each experiment, 2 mL mixed Iiquor samples were periodically withdrawn
From the serurn bottles. The sample was split into two I mL aliquots, one for VFAs
measurement and the other for measurement of LCFAs. Deionized water (2 mL) was
used to dilute the VFA aliquots. Afier centrifugation at 1750 g for 5 minutes, the centrate
was removed. filtered through H-cartridges (Dionex Canada) and diluted with deionized
water to maintain a detector response within the range of the caiibration curves. The
filtered samples were analyzed using a Dionex Ion Chromatograph equipped with a 25
pL sarnple loop. a conductivity detector (CD 20). a 24-cm x 4-mm diameter AS 1 1
colurnn. an AMMSII micromembrane supressor and a lon~ac" ATC-1 cartridge (al1 fiom
Diones). The three eluents used were deionized water (eluent A), 5 mM NaOH (eluent
B) and 50 mM NaOH (eluent C). The eluent flows as a percent of the total flow of 2
ml-min" were as follows: O - 2 mins.. 93% A. 7% B: 2 - 6 mins.. A ramped from 93% to
0%. B from 7% to 100% 6 - 9 mins., B ramped from 100% to 50%. C ramped from 0%
to 50% and then held unti19.99 minsj and fiom 10 - 17 mins., 93% A, 7% B. This
rnethod provided detection of acetic (CI) acid. propionic (C;) acid. iso- and n-butyic (Cr)
acids. iso- and n-valeric (Cj) acid and n-hexanoic (C6) acid. The effective detection
limits (incorporating dilution) were 0.2 r n g ~ - ' for propionic (C3). i -butyric (Cr). n-
valeric (Cs) and i -valeric (C6) acid: 0.3 r n g ~ - ' for acetic (C2) and n-butyric (Cr) acid and
0.4 mg-l" for hesanoic (Cs) acid.
TripIicate standards for VFA analysis were prepared in basal medium using a 1500
mg^" VFAs stock solution. The stock solution was prepared with acetic (Cz) acid.
ï h e first criteria for choosing a dispersing agent were based on polarity and its ability
to solubilize LCFAs. Solubility data for LCFAs in ethanol is unavailable. However.
based on their solubility in acetone (a polar solvent) and n-heptane (a non-polar solvent)
(Bagby. 1993). some solubility is expected in ethanol. As a cosubstrate, the behavior of
ethanol was of concem because of its degradability and potential influence on LCFA
degradation. Ethanol could enhance LCFA degradation. Although LCFA solubilities in
acetyl acetate are unknown. it was considered as a dispersing agent because of its
polarity: however. concerns for its use were similar to those for ethano1 and it was
discarded as an option.
Ethers and esters are less polar in comparison to alcohols bearing the sarne number of
carbons (Smallwood, 1996). Ethers also have an advantage of being recalcitrant (Yeh
and Novak. 1994). Although the alcohol and ester were initially rejected as potential
dispersing agents. they were evaluated with the two ethers for their toxicity effect using
data from the literature. For the tosicity evaluation comparison. similar compounds were
esaminsd using data reported by Playne and Smith (1983) since comparative microbial
toxicity data for ethanol. ethyl acetate. diethyl ether and MTBE are unavailable. Playne
and Smith (1 983) reported toxicity. measured on a methane production basis. for isoamyl
alcohol. isoamyl ester and isoamyl ether (2.5 % (v/v)) to be approximately the same.
These researchers esamined the toxicity of several chernicals used for product extraction
from anaerobic process streams. Using data reported by Playne and Smith (1983),
diethyl ether and MTBE were identified as candidates for delivering LCFAs. In
comparison to diethyl ether, MTBE was more toxic (Lewis, f997) for use in the
laboratory and was eliminated as an option.
Two-mL of diethyl ether was the maximum amount found to be effective in dispersing
10 mg of C 1 8 LCFAs (1 0 mg per 100 mL of liquid). the maximum amount used in this
study. The concentration of LCFA stock solution used through out this study was 5000
r n g - ~ - ' . In control cultures, acetate was used as a measure of diethyl ether hydrolysis and
insignificant Ievels were detected. In cornparison to controls without diethyl ether. no
increase in methane levels were observed in control cultures receiving diethyl ether.
3.7 Long Chain Fatty .4cid (LCFA) Measurement
3 -7.1 LCFA Extraction- Method Development
LCFAs are relatively insoluble in aqueous solutions and an extraction protocol was
developed to ensure losses did not occur afier samples were removed frorn the serum
bottles. Solvent polarity. pH. ionic strength and extraction time were investigated to
determine their effects on the arnount of LCFAs recovered during liquid sarnple
extraction. Several organic extractives Lvere screened to compare solvent polarity and
extraction efficiency: chloroforrn. chloroform:methanol (1 : f ). hexane and hexane:MTBE
( 1 : l ) .
-411 samples for extraction studies were prepared in triplicate using 10 mL of an
anaerobic culture in 20 mL serum bottles. The bottles were sealed ~ ~ i t h TeflonB fined
septa. secured with alurninum caps and shaken using an orbital shaker (Lab Line
Instruments Inc. Mode1 No. 3520) for 10 minutes at 200 rpm. One-mL sarnples were
removed and transferred into 5 mL semm bottles with 2 mL of an organic extractive.
The extraction time was initially set at 5 minutes for a11 the extractives. However. for the
hexane:MTBE extractive. increased extraction effrciencies were observed when
comparing results for the 5 and 15 minute extraction times. Afier shaking. the samples
Lvere centrifuged for 5 minutes at 1750 g to separate the aqueous and organic layers.
One-pL samples of the organic phase were analyed by gas chromatography.
Chloroform. a polar solvent- \vas initially examined to determine the percent LCFAs
recovered afier extraction (Figure 3.1 ). Percent recoveries for dodecanoic (C 10) to
octadecanoic (C 18) LCFAs were greater than 90 percent for cultures receiving 56.25 and
1 2 . 5 m . However. l o w r LCFAs extraction efficiencies between 70 to 85 percent
nvre obtained for cultures receiving 5.625 r n p - ~ - ' . Also. less than 50 percent hesanoic
(C6) and octanoic (Cs) acids were recovered from cultures receiving 5.635 and 56.25
mg-L-'. To increase the amount of LCFAs recovered from cultures receiving
approsimate1)- 5 n q - ~ - ' . a 1 : I misture of rnethanol and chloroform was used and
recoveries increased to between 80 to 95 percent. Usine the methanoI:chloroforrn
mixture also improved percent recoveries for hesanoic (C6) and octanoic ( C g ) acids at ail
concentrations esamined (Figure 3.2). While an effective extractive. chlorofonn \vas
considered too tosic to be used in the laborator). on a long-term basis. As a result.
alternative sol\-ents were investigated based on data obtained from the
n~etl~ano1:chloroforrn study.
Hesane. a non-polar solvent. \vas the next extractive examined and LCFA extraction
efficiencies were found to be similar to chloroform (Figure 3.3). Percent recoveries were
comparable to chlorofonn for cultures receiving 56.25 and 1 12.5 r n g . ~ - I . However, at
5.625 mgLe ' . less LCFAs were extracted into the hexane phase in comparison to the 1 :1
methanol:chloroform misture.
Figure 3.1 : Percent LCFA extracted into chloroform (Averages for triplicate samples, error bars represent standard deviation for the samples)
Figure 3.2: Percent LCFA extracted into ch1oroform:methanol ( Averages for triplicate sarnples, error bars represent standard deviation for the samples)
Figure 3 . 3 . Percent LCFA extracted into hexane (.\verages for triplicate samples, error bars represent standard deviation for the sarnples)
I Herane:31TB E with SaCI and pH LCFA O Hcxanc O Heranc:MTBE
Figure 3.4: Percent recovery for 5.625 mg-^^' C6 to Ci* in hexane and hexane:MTBE: 5 minutes shaking at 200 rpm (Averages for triplicate samples, error bars represent standard deviation for the sampies)
IIIcrane: . \ ITBE with S ~ C I and pH U l iesane LCFA O Hexane:>lTB E
Figure 3.5: Percent recovery for 5.625 mg^-' C1 to Cis in hexane and he'rane:MTBE: 15 minutes shaking at 200 rpm (Averages for triplicate samples, error bars represent standard deviation for the samples)
Several parameters were investigated to increase the amount of LCFA extracted into
the hexane phase. Longer extraction times fiom 5 to 15 minutes assisted in increasing the
amount recovered for cultures receiving 5 -625 mg-^-' LCFAs (Figures 3.4 and 3.5).
Adding a 1 : 1 mixture of MTBE and hexane, lowering the pH (2 drops of 1 : 1 H2S04) and
adding NaC1 (0.05 g) were examined to fùrther increase percent recoveries (Figure 3.5).
Greater than 95 percent CIO to Ci* LCFAs were extracted into the hexane:MTBE phase at
lower pH values and increased ionic strength.
For octanoic (Cg) acid. the incremental amount recovered at longer extraction times
increased by approximately 25 percent and minor increases were observed for hexanoic
(CG) acid. The final extraction protocol developed for the remainder of this study
consisted of adding 0.05 g NaCl. 2 drops of 50 % H2S04. 2.0 mL of a 1 : 1 mixture of
hesane and MTBE to ImL aqueous samples in 5 mL serum bottles. The bottles m-ere
shaken using an orbital shaker at 200 rpm for 15 minutes and centrihged for 5 minutes at
1750 g to separate the organic and aqueous phases. One-pL samples were removed from
the organic phase and analyzed by gas chromatography.
This method quantified saturated LCFAs (dodecanoic (Clo to stearic (Cl8 acids
and unsaturated C 18 LCFAs (linoleic (Ci8 ,) and oleic (Cl* 1 ) ) with a 85 to 90 percent
removal efficiency. Octanoic (Cs acid was characterized with a 75 percent removal
efficiency.
3 -7.2 LCFA Extraction- Phase Partitionhg Studies
LCFAs are biodegraded only when they adhere ont0 LCFA degrading organisms.
Because Cis LCFAs have only limited aqueous solubility (2.9 mg^-'). it was n e c e s s e
to examine if the LCF.4.s partitioned ont0 the glass surface or with the biomass. Serurn
bottles (20 mL) were inoculated \vith 10 mL of 1500 m g ~ ~ ~ - ~ - l biomass and kno\r-n
quantities of linoleic (Cls,z) acid \vas added from a 5000 r n g - ~ - ' diethyl ether stock
solution. Afier feeding the cultures with 100 r n g ~ - ' linoleic ( C [ S . ~ ) acid, 3-3 mL samples
Lvere removed from the serum bottles and split into 1 mL (sarnple A) and 2 mL (sample
B) portions. Sarnple B was centrifuged and 1 mL of the centrate removed and placed into
a 5 mL via1 (Sample C). The amount of LCFAs in samples A (culture + basal medium)
and C (basal medium) was determined using the extraction procedure outlined in section
3.7.1 . The average linoleic (C 8 2 ) acid concentrations distributed within the microbial
culture (sample A) and centrate (sample C) were 100 + 5 r n g * ~ - l and 5.8 + 0.6 mg^-'.
respective1 y (Figure 3 -6) . The slightly higher linoleic CI^:^) acid concentration (5.8 i 0.6
decrease (Figures 5.3 and 5.1) and large quantities of acetate of up to 200 r n g - ~ - '
accumulated in cultures receiving 50 and 100 r n g - ~ - ' linoleic (CIg:2) acid. At I O mg-L-' - linoleic (Cig:-) acid, most of the acetate was removed by day 70. Trace quantities of
propionate (C3), iso- and n-butyrate (C4) and iso- and n-valerate (CS) were detected at dl
linoleic (C :) concentrations examined (data not shown).
-
O 20 4 0 60 80 100 120 1 40
Time, days
Figure 5.6: Effect of linoleic (CIg:2) acid (LA) concentration on acetate production. (Average for triplicate samples.)
\,lethane \vas produced at al1 linoleic (C18.~) acid concentrations esarnined (data not
shown). Mass balances shown in Figure 5.7 were calculated by convening al1 the
byproducts (escludin_g methane) to both a linoleic (CIR:~) acid basis and a carbon basis for
cultures receivine - 10, 50 and 100 r n g - ~ - ' linoleic ( C i g : ~ ) acid (see example calculations in
Xppendis B). AIthough fluctuations were noted, in panicular for cultures with 100
r n g - ~ ' ' linoleic (CIO:2) acid, the mass balances indicate that the primary products of
linoIeic acid degradation were detected and measured.
Figure 5.7: Linoleic (C18.2) acid degradation study mass balance. (Averages for triplicate samples. error bars represent standard deviation for the samples)
5.1.2 Inhibiton Effects of Linoleic (CI8:?) Acid on Methanogenesis
-4. -4 cerare Degradation
Acetate degradation was examined using 1600 i 100 m g L-' VSS in the presence and
absence of diethyl ether for duplicate control cultures (Figure 5.8). Within approximately
2 da!.s. cultures containing neither linoleic (CI8.?) acid nor diethyl ether consumed the
added 100 mg^-' acetic acid. Acetate degradation was inhibited and complete
consumption was accompIished within approximately 14 days in the presence of diethyl
ether. although acetic acid was 80 to 90 % removed within 2.5 days (Figure 5.8).
Acetate degradation profiles for cultures receiving Iinoleic (Cl8 -) acid are shown in
Figure 5.9. No inhibition was observed when comparing cultures receiving only diethyl
ether uith those receiving 10 mg^" linoleic (Cigi2) acid. However, at concentrations of
Up to 10 days after inoculation, none of the initially added acetate was removed in one
culture receiving 100 rng.~- ' linoleic (Clg:Z) acid (Figure 5.9, test :! 1). In a later study
( 100 mg^" linoleic acid test R), aceticlastic methanogenesis was inhibited up to
day 15 in comparison to the control cultures. M e r day 10, however, the inhibition was
relieved in one culture (test ff2) but remained in the other culture even at day 25 (test g1).
Initial acetate degradation rates for control cultures and those receiving 10 mg^'
linoieic (Clw -) acid are sho\\-n in Table 5.2. In comparison to control cultures, diethyl
ether reduced the acetate degradation rate by approximately 77%. The presence of I O
r n g - ~ " linoleic acid decreased the initial acetate degradation rate by approximately
30°% in cornparison to the diethyI ether control.
O 0.5 1 tS 2 25
-w Figure 5.8: Acetic acid removal in the absence of linoleic acid (LA).(DE = diethyl ether, average for duplicate results)
+ 10 mglL LA test 2
-30 mglL L A test 1
- 5 0 mglL LA test 1
e l 0 0 mg/L LA test 1 - I I - + 100 mg/L LA test 2
O 5 10 15 20 25 30 35 Time, days
Figure 5.9: Acetic acid removal in the presence of linoleic (C18:2) acid (Average for tnplicate samples shown)
1 Average of duplicates, ' ~ v e r a ~ e and standard deviation of triplicate samples. DE = diethyl ether, LA = linoleic (CIK:t) acid, ND = Not degraded
i. Hydrogen uptake 1 day afier linoleic (Clg:~) acid addition
Hydrogen uptake profiles shown in Figure 5.10 are for duplicate control cultures in
the presence and absence of diethyl ether. Al1 experiments were conducted using 1550 k
90 r n g - ~ " VSS. The differences in rate constants observed among control cultures are
Condition Esamined
Control Control + DE 1 O mg^" LA 30 r n g - ~ " LA 50 rng.~" LA
Initial Acetate Deeradation Rates ( mg~-'-d-') 2
85 1 9'
13.6 k 0.3' 2
-
ND
ND A
ND
negligible indicating that diethyl ether caused insignificant inhibition of
hydrogenotrophic methanogens. At the four concentrations examined, tinoleic (CIR:P)
acid did not significantly inhibit hydrogenotrophic rnethanogenesis and hydrogen was
consumed within 12 hours (Figure 5.1 1).
The shape of the hydrogen uptake profiles was non-linear and followed a first order
kinetic expression (equation 5.2, where Hz is the hydrogen concentration (pmoles /
bonle) and k is the first order rate constant, d-'). The first order expression was
used to compare the data sets using least-squares regression.
The hydrogen first order rate constants were detennined for the data in Figure 5.1 1
and are shown in Table 5.3 . NI data sets provided rZ > 0.97 for the least-squares
Fisure 5 . IO: Hydrogen removal in the absence of linoleic (Ci rr:2)
acid (LA). (DE = diethyI ether, Average for duplicate results shown)
Figure 5.1 1 : Hydrogen removal in the presence of linoleic (Cis:2) acid (LA). Experirnent performed 1 day after L.4 inoculation. (Averages for tri plicate samples, error bars represent standard deviation for the samples)
regession. Based on the Tukey's paired cornparison procedure (Box et al.. 1978) at a 95
"io confidence level, the rate constants for cultures receiving O (without diethyl ether), O
(with diethyl ether) and 10 r n g - ~ - ' linoleic (Cig,z) acid were not significantly different
from each other but were significantly different from those for cuitures receiving 30, 50
and 100 rne .~ ' ' linoleic (Cin,2) acid. No significant differences were found between rate
constants for cultures receiving 30, 50 and 100 r n g ~ - ' linoleic acid.
i i . Hydrogen uptake 18 days and 35 days afier linoleic ( C l g , ~ ) acid addition
Cultures. inoculated with linoleic (Clg,2) acid and receiving hydrogen one day after
preparation of the serum bottles, were injected with hydrogen again on day 18 and day
35. Data for hydrogen profiles shown in Figures 5.12 and 5.13 for day 18 and day 35,
respectively were used to caiculate first order constants for hydrogen uptake rates shown
in Tables 5.4 and 5.5. Statistical differences between rate constants for dôy 18 and day
3 5 were evaluated using the Tukey's paired cornparison procedure (Box et al., 1978).
Table 5 - 3 : First order rate constants for hydrogen removal 1 day afier adding LA
' LA Concentration, rng-L1 First Order Rate Constant, d"
Average andstandGd deviation for triplicate samples shown DE = diethyl ether, LA = linoleic (Cie:2) Acid
O (without DE) O (with DE)
10
6 l ime , hrs
0.3871 0.006 0.3791 0.006 0.374 I 0.006
Figure 5.12: Hydrogen removal in the presence of linoleic (Clg2) acid (LA) 18 days afler LA addition. (Control duplicates; Average for triplicate samples and the error bars are standard deviation for the samples)
Significant differences were found between the rate constants s h o w in Tables 5.4 for
cultures receiving 10 and 30 r n g - ~ " linoleic (C1gZ2) acid in cornparison to those cultures
receivine 50 and 1 00 mg^" linoleic (C s:z) acid. Uptake rate constants for cultures
Figure 5.13 : Hydrogen removal in the presence of linoleic (Cls:~) acid (LA) 35 days afier LA addition. (Control duplicates; Averages for triplicate samples, error bars represent standard deviation for the samples)
Table 5.4: First order rate constants for hydrogen removal 18 days afier adding LA
1 Duplicates, ' ~ v e r a ~ e and standard deviation for tnplicate samples L.4 = linoleic (C1g2) k i d .
LA Concentration, rng -~~ '
O (without DE) O (with DE)
10 30 50 1 O0
First Order Rate Constant, d"
1
0.4 19' 0.396~ ,
0.374 t 0.007~ 0.359 i 0.001~ 0.323 t 0.004~ 0.324 + 0.009' .L
Table 5.5 : First order rate constants for hydrogen removal 3 5 days after adding LA
receiving 50 and 100 mg-^-' linoleic (Cis:z) acid were significantly different from those
L.4 Concentration. r n g - ~ - '
O (without DE) O (with DE)
1 O 30 50 1 O0
receiving 30 mg^" linoleic CI^,^) acid. In comparison, no significant difference was
First Order Rate Constant, d"
0.384~ 0.364'
0.337 I 0.006~ 0.332 I 0.0003~ 0.3 18 A 0.003' 0.330 I 0.006~
obsenxd for cultures receivine 50 and 100 r n g ~ - ' linoleic (Cin:2) acid At day 35.
' ~ u ~ l i c a t e s ~ ' ~ v e r a ~ e and standard deviation for triplicate samples L.4 = Iinoleic (C18:2) Acid
significant differences between the uptake rate constants were not observed at any of the
linoleic (C18:') acid concentrations examined. Duplicate cultures with and without
diethyl ether Lvere not compared because the Tukey's paired comparison procedure
accumulation was limited and was degraded to undetectable ieveis within approximateiy
12 and 28 days in cultures receiving 30 r n g - ~ - l and 50 rng-L-' oleic (Ci8:i) acid,
respectively. Similady, myristic (Clr:~) acid (Figure 6.3) accumulated to 10 mg-^-' in
cultures receiving 50 r n g l - ' and 100 mg-L-' oleic (Cls,i) acid. No stearic (Cls:o),
paimitoIeic ( C I ~ J ) , lauric (C12:0) nor hexanoic (C6) acids were detected at any oleic (CM:~)
acid concentration examined.
In cultures receiving 10, 30 and 50 r n g - ~ " oleic (Clarl) acid (Figure 6.4), less than 10
r n g ~ - ' of acetate was detected. However, acetate accumulated between 20 to 30 days but
was undetected within 45 days in cultures receiving 100 r n g - ~ - ' oleic acid.
Mass balances are shown in Figure 6.5 for al1 oleic (Cie,l) acid concentrations
esamined. The baIance is based on a sum of oleic (Cis:r) acid, LCFA fi-oxidized
O 5 10 15 20 25 30 TI me, days
Figure 6.2: Palmitic (&:O) acid profiles in cultures receiving oleic (C1n:i) acid (OA). (Average for triplicate samples.)
Figure 6.3: Myristic (Cl4:o) acid profiles in cultures receiving oleic (Ci ,) acid (OA). (Average for triplicate samp les.)
O 10 20 30 40 50
Time, days
Figure 6.4: Effect o f oleic (Ci8:1) acid (OA) concentration on acetate production. (Average for triplicate samples.)
byproducts plus acetate and is reponed on an oleic (C18:1) acid and carbon basis (see
exarnple calculations in Appendix B). Although fluctuations are seen in Figure 6.5, the
profiles account for the initially added oleic acid and al1 the byproducts of P-
osidation.
6.1.3 Inhibitoq Effects of Oleic (Cis:r) Acid on Methanogenesis
In cornparison to controls without diethyl ether, acetate degradation was inhibited in
the presence of diethyl ether (Figure 6.6). Al1 experiments were conducted using 1 500 rir
70 r n g - ~ - ' VSS. Profiles in Figure 6.7 show acetate degradation was also inhibited by
oieic (C,SJ) acid at al1 concentrations examined. Although inhibition was observed at 10
Time, days
Figure 6.5: Oleic (CIR:I) acid degradation study mass balance. (Averages are for triplicate samples, error bars represent standard deviation for the samples)
O 5 10 15 20 25
liffi*~ Figure 6.6: Acetic acid removal in the absence o f oleic ( C l g : ~ ) acid. (OA = oieic (CIR:I) acid, DE = diethyl ether, average for duplicate results)
Figure 6.7: Acetic acid removal in the presence o f oleic (Clg:~)
acid (OA). (DE = diethyl ether; Duplicate control; Average for triplkate samples)
rn=-~-' oleic (Cip;I) acid, acetate was undetected between day 25 and day 30. In
comparison. acetate accumulated and was not degraded in cultures receiving 100 rng~-l
oleic (C 18: 1) acid up to day 25.
Initial acetate degradation rates are s h o w in Table 6.1. In the presence of diethyl
ether, the acetate degradation rate was reduced by approximately 85% in comparison to
control cdtures. When comparing cultures receiving diethyl ether, oleic (C18:1) acid also
influenced aceticIasric methanogens at 10 r n g - ~ - ' as a 60% rate reduction was observed.
Table 6.1 : Initial acetate degradation rates for varying oieic (C is : , ) acid
B. kiydmger~ Cotmcmpriorr
Profiles for control cultures without the presence of oleic (Clg:~) acid in Figure 6.8
sholv that hydrogen removal was not affected by the presence of diethyl ether. Al1
esperiments were conducted using 1600 i 80 r n g - ~ " VSS. In the presence of oleic
(C acid, small differences were observed between profiles for cultures receiving 10,
30. 50 and 100 mg^-' oleic (Cig,l) acid. Hydrogen uptake rate constants shown in Table
6.2 were calculated assuming first order kinetics. Using the Tukey's paired comparison
procedure (Bos er al., 1978). the first order rate constants for al1 the oleic (C18:[) acid
concentrations h
Condition Examined
Control Control A DE 10 r n g ~ - ' 0.4 3 0 mg^-' 0.4 50 mg-^" OA 1 00 mg L-' O A
Initial Acetate De radation Rates S ( mg-L- *d-') 54' 8'
3 -C 1' ND ND ND
Duplicates, 'r2verage and standard deviation for triplicate sarnples. ND = Not degraded, DE = diethyl ether, OA = oleic (Clgl) acid.
concentrations examined were compared. The constants for cultures receiving 30, 50 and
100 r n g - ~ - ' were not signiticantly different fiom each other but were aatistically
Figure 6.8: Hydrogen removal in the absence of oleic (Cis:~) acid (0.4). (DE = diethyl ether, Average for duplicate results shown)
Figure 6.9: Hydrogen rernoval in the presence of oleic (Clg: , ) acid (OA). (Averages for triplicate samples, error bars represent standard deviation for the samples)
different (95 % confidence) frorn those receiving IO mg^"
Table 6 2: First order rate constants for hydrogen removal in the presence of oleic (C,g:l) acid
6.2 Discussion of Results
6.2.1 Oleic (C 18: 1 ) Acid Degradation
By combining the results presented in Chapter 5 with those presented in this chapter
the question of whether hydrogenation of unsaturated Ci* LCFAs takes place prior to P-
osidation can be addressed. During linoleic (Clg12) acid degradation, oieic (CI 8. l) acid
[vas formed as a trace product (Figure 5 . 9 , indicating that hydrogenation could occur
prior to P-osidation. However. the results in Figure 6.1 show that oleic (Cls:~) acid was
degraded within the same time period as linoleic acid. This indicates that
hydrogenation of iinoleic (CIR:2) acid was not required prior to P-oxidation. If
hydrogenation of linoleic (CIsz2) acid was a necessary step prior to P-oxidation, oleic
(Cis 1 ) acid should have accumdated to rnuch higher concentrations, to the order of the
initial linoleic acid concentrations, assuming the rate of conversion is sfow. Based on the
results presented in Chapter 5 and those obtained frorn studies with oleic (Cl*:,) acid, it
Oleic (Clg:~) Acid Concentration, mg L-'
O (without DE) O (with DE)
10 30 50
1 O0
appears that complete saturation was not required before P-oxidation since no stearic
First Order Rate Constant, hf'
0.300' 0.295'
0.262 I 0.004' 0.222 i 0.008' 0.209 k 0.001' 0.212 ,+ 0.003'
' ~u~ l i ca tes ' ' ~ v e r a ~ e and standard deviation for tnplicate samples DE = diethyl ether
(Clg) acid \:as detected as a by-product of either linoleic (Cl&?) or oleic CI^:^) acid. This
observation fûnher suggests that complete saturation was not required. Funher support
for this hypothesis is provided by observation of palmitoleic (Cl6:I) acid (Figure 5 .3 )
fiom linoleic acid experiments.
Although complete hydrogenation does not appear to be a necessary step prior to P-
osidation. the data from linoleic (C18:2) and oleic (C18:1) acid degradation support the
hypothesis that hydrogenation and P-oxidation may occur concurrently. The formation of
palmitic (Cl6.0) and palmitoleic (Cl6.1) s i r n u l t a n e ~ ~ ~ l y (Figure 5.3) supports this point of
view .
The primary byproduct of oleic (Clgri) acid degradation was the h l l y saturated
palmitic (C16:O) acid (Figures 6.2). If P-osidation occurred before hydrogenation,
unsaturated Clo LCF.4s should have been observed. The possibility exists that
unsaturated C16:1 LCFAs were produced but then were rapidly converted to the relatively
slou.ly degrading palmitic (C16:o) and myristic acids. Studies with saturated C16:0
and unsaturated C16;1 LCFAS would be required to test this hypothesis. In addition to
palmitic (C 16:0) and myristic (CIJ:~) acids, Canovas-Diaz el al. (199 I ) also observed the
presence of palmitoleic (C16.1) acid during anaerobic conversion of oleic (Ci8 ,) acid.
In previous studies by Novak and Carlson ( 1 970), only trace amounts of Ciz:~ to Cg:O
acids and no acids were observed during degradation of linoleic (Cle:~), oleic (C18:1),
stearic (C 1 pdmitic (C i6:0) and rnyristic (C i4:o) acids. However, in cornparison,
significant concentrations o f C16:o and Cr+o acids were detected fiom oleic ( C I R : ~ ) acid
depadation. No Cg and C I O acids were observed at any concentration o f the oleic (Cigl)
acid examined. These observations support the hypothesis that f3-oxidation and not a- or
w- oxidation is the prirnary mechanism for Cl8 LCFA degradation in anaerobic systems.
Accumulation of palmitic (&:O) and myristic (C1.4 acids during the degradation of
50 r n g - ~ - ' and 1 00 rng-~- ' oleic (C 18:~) acid (Figures 6.2 and 6.3) suggested that these
byproducts may be inhibitory to the microbial cornmunity. However, inhibitory effects
of organisms consuming palmitic (&:O) and myristic ( C I J : ~ ) acids were eventually
relieved as these acids were degraded and hence. removed fiom solution. In cornparison.
linoleic (CIK:f) acid caused a greater residual inhibition of its byproducts than did oleic
(Cix:l) acid as the inhibition was relieved sooner, afier approxirnately 5 to 25 days in
oleic CI^:^) acid fed cultures.
The mass balance accounted for al1 the initially added oleic (Cig:~) acid and its B-
osidized byproducts up to day 15. The decrease observed after day 15 is due to the
consumption of acetate.
6 .2 .2 Oleic ( C I X : ~ ) acid-Methanosenic Inhibition Studies
Similar to studies conducted with linoleic ( C I R : ~ ) acid, no evidence was found for
acetate inhibition caused by synergistic effects between diethyl ether and oleic (CI*: 1)
acid. Inhibition of acetate consumption was observed in cultures receiving greater than
I O r n g ~ - ' oleic (Cla:l) acid. In cornparison, Angelidaki and Ahring (1992) and Koster
and Cramer (1987) reponed greater than 300 rng-~-' oleic (Clg:~) acid inhibited
consumption of acetate. Differences in concentration at which inhibition is initiated in
this work and that reported by Angelidaki and Ahring (1 992) and Koster and Cramer
( 1 987) are probably due to culture adaptation as welI as the high temperature conditions
of 55 OC and 30 OC respectively, under which the latter studies were conducted.
Additional differences may be due to the VSS concentration used in this study in
comparison to those reported by Angelidaki and Ahring (1992) and Koster and Cramer
(1987) in their studies. Angelidaki and Ahring (1992) reported al1 experiments were
conducted with 5% v/v cattIe manure instead of VSS concentrations. Koster and Cramer
(1 987) did not report VSS concentrations in their study.
Methanogenic inhibition by milk fat containing oleic (Cis;l) acid was also reported by
Perle el al. ( 1995). Although acetictastic or hydrogenotrophic methanogenic inhibition
was not specifically investigated, they reported significant methanogenic activity losses
for cultures inoculated with milk fat or oleic ( C I * : ~ ) acid.
6.3 Summary
Oleic (Cl*:,) acid was degraded to shorter chain LCFAs and ultimately to acetic acid at
2 1 "C. During oleic (Cisyi) acid degradation, intermediate length Cio to Cg fatty acids
were not obsewed as P-osidation degradation byproducts at any concentrations
examined. In comparison to studies with iinoleic ( C I R : ~ ) acid, palmitoleic (Cl6 1). lauric
( C 4 and hexanoic (C6) acids were not detected dunng the degradation of 100 r n g ~ - '
oleic (C l x . 1) acid. Oleic (Clx:~) acid was degraded to pairnitic (C ic:o), myristic (C14:0) and
acetic acids at 2 1 "C. LCFA byproduct distribution profiles were a function of the oleic
(Clx:~) acid concentration. Acetic acid did not accumulate in cultures receiving less than
100 mg-L" oleic (Cia:l) acid during the degradation studies.
Diethyl ether inhibited aceticlastic methanogens but not hydrogenotrophic
methanogens. Aceticlastic methanogens were affected by oieic (C1s:1) acid at al1
concentrations examined. At greater than 10 rng-~-' oleic (C 18:~) acid, aceticlastic
methanogens were initially inhibited but the inhibition was relieved after approximately
10 days. Although hydrogenotrophic methanogenesis was inhibited, the inhibition was
minimal.
7.0 DEGRADATION OF STEARIC (Ciri:o) ACID AND ITS lNHIBITORY EFFECTS ON METHANOGENESIS
7.1 Experimental Results
7.1 - 1 Stearic (Cl s:o j Acid Degradation
Steanc (C~~NI) acid degradation profiles at 2 1 O C are shown in Figure 7.1. Al1
esperiments were conducted using 1600 i 80 rng -~ ' ' \%S. .Mer 55 days, approsimately
50 to 60 % of the stearic acid remained undegraded in cultures receiving 10, 30.
50 and 1 00 rng-~-'. No LCFA byproduas were observed, however, acetate was detened
at al1 stearic (&O) acid concentrations examined (Figure 7.2). Acetate accumulated to
between 40 to 50 rng~-l and was degraded after day 40 only in cultures dosed with 10
rng~-l. However, in cultures receiving 30, 50 and 100 r n g - ~ " stearic (C1s:o) acid, acetate
was not degraded. Degradation rates calculated from slopes of the curves in Figure 7.1
are shown in Table 7.1 for al1 stearic acid concentrations examined. The rates are
s h o w to increase with the stearic acid concentration from 10 to 100 rng~'~.
A mass balance for al1 the stearic (CIx:o) acid concentrations examined is shown in
F ip re 7 .3 . The balance is based on convening the acetate byproduct to a stearic (Cla:o)
acid basis and adding it to the amount of measured stearic (Cig,o) acid. .A carbon balance
is also presented in Figure 7.3 (see example calculations in Appendix B).
Figure 7.1 : Stearic (C18.0) acid (SA) degradation profiles in cultures with stearic (Ci* acid. (Average for tnplicate samples.)
O 10 20 30 40 50 60 lime, &ys
Figure 7.2: Acetate production profiles for cultures fed with stearic (C 1 *:O) acid (SA). (Average for triplicate samples.)
Figure 7.3 : Stearic (C s:o) acid degradation mass balance. (Averages for triplicate samples, error bars represent standard deviation for the sampIes)
7 1.2 Inhibiton. Effects of Stearic (Cig:o) Acid on Methanogenesis
A. A cerate Degradation
Diethyl ether and stearic ( C i s ; o ) acid inhibited methanogenesis at al1 concentrations
examined (Figures 7.4 and 7.5). In comparison to controls, diethyl ether reduced the
acetate degradation rate by approximately 87%. Unlike linoleic (Cia,2) and oleic (C18:l)
acids. however, increasing concentrations of stearic (C1r:o) acid provided no additional
inhibition. Initial acetate degradation rates for control cultures and for those receiving
stearic (Cis:o) acid are shown in Table 7.2. The acetic acid degradation rate for al1 stearic
(CIX.O) acid concentrations was approximately 45 % of the degradation rate in the diethyl
ether controls (Table 7.2). During the duration of the experiment, stearic (Cis:o) acid was
not completely degraded and remained in the system. Based on the Tukey's procedure,
Table 7.2: Acetate degradation rates for varying stearic acid
Figure 7.4: Acetic acid removal in the absence of stearic ( C i s : ~ ) acid (SA). (S.4 = stearic ( C i s : ~ ) acid, DE = diethyl ether, average for duplicate results)
concentrations
there are no significant (95 % confidence) differences observed between the rates at any
of the concentrations examined. Al1 experiments were conducted using 1500 + 100
Condition Examined
Control Control + DE 10 r n g ~ - ' SA 30 r n g ~ - ' SA 50 r n g ~ - ' SA
3
Acetate Degradation Rates ( mg^" -d-')
1
61 -8' d
7.90' 3.66 î 0.23~ ,
3-50 I 0.30' 3.32 1 0.13'
1 00 r n g ~ - ' SA w
3.78 I 0.33' A
T~uplicates, '.Average and standard deviation for uiplicate samples DE = diethyl ether. SA = stearic (&:O) acid
Figure 7.5: Acetic acid removal in the presence of stearic (CI8:O) acid (S.4). (DE = diethyl ether; Duplicate control; -4verage for tnplicate samples.)
Comparing control cultures, the presence of diethyl ether had very little effect on
hydrogenotrophic methanogens and most of the hydrogen was removed within 12 hours
(Figure 7 . 6 ) . Similarly, profiles for cultures dosed with 10, 30, 50 and 100 r n g ~ - ' stearic
(CIR:O) acid show most of the hydrogen was removed within 12 hours (Figure 7.7). Rate
constants for hydrogen uptake were calculated assuming first order kinetics and are
shown in Table 7.3. Using the Tukey's procedure to compare the stearic (Cls:o) acid first
order data indicated that there were significant (95 % confidence) differences observed
between the rate constant for cultures receiving 10 rng-~- ' stearic (Clsa) acid in
cornparison to those receiving 30, 50 md 100 rng-~ - ' stearic (Cis:o) acid. No significant
A no SA. no DE
r m SA. with DE
Figure 7.6: Hydrogen removal in the absence o f stearic (CIK:O) acid (SA). (DE = diethyl ether, Average for duplicate resuits shown)
O I> 7 4 6 8 10 12
Time, hrs
Figure 7.7: Hydrogen removal in the presence o f stearic (Cig:o) acid (SA). (Averages for triplicate samples, error bars represent standard deviation for the samples)
differences were observed for cultures receiving 30, 50 and 100 rng-~-' stearic (Cls:a)
acid. hl1 experiments were conducted using 1500 + 90 mg^" VSS.
I Duplicates, '!iverage and standard deviation for tnplicate samples
DE = diethyi ether
Table 7.3 : First order rate constants for hydrogen removal in the presence of stearic (C I~MI) acid
7.2 Discussion of Results
7.2.1 Stearic (Cix.o) Acid Degradation
hlackie er al. ( 199 1 ) and Novak m d Carlson (1 970) proposed that complete double
bond saturation is necessary to initiate the P-oxidation. However, our results show
stearic (Cls:~) acid to be less degradable than linoieic (CIR:~) and oleic (CIR,~) acids.
Therefore. the hypothesis that stearic (Clx:") acid is the only Cl8 acid to initiate P-
Stearic (C l*:~) Acid Concentration, r n g ~ - '
O (without DE) O (with DE)
10 30 50 1 O0
i
osidation is unlikely. Support for P-oxidation of unsaturated LCFAs is evident by work
conducted during this study and that by Canovas-Diaz et al. (199 1) with the formation of
palmitoleic acid from linoleic CI^:^) acid (chapter 5 of this work) and from oleic
(C acid, respectively.
Degradation - rates shown in Table 7.1 increased with stearic (Clg:~) acid concentration.
Lower degradation rates observed for steuic (C18:0) acid in cornparison to linoleic
and oleic (C lg:~) acids may be due to slow uptake of stearic (C1a:o) acid into the cell. The
First Order Rate Constant, d-'
m
0.385' 0.38 1 '
0.378 + 0.012~ ,
0.362 I 0.017' 0.341 k 0.008' 0.342 +, 0.004'
biochemicai reason for observing low stearic (Cls:~) acid degradation rates is unclear, but,
it is possible that the formation of a stearic (Clg:~) acid acetyl CoA complex to initiate the
P-oxidation pathway is the rate determining step.
Palmitic (CIG:O) and myristic (C14:o) acids. detected during the degradation of linoleic
(C ls:z) and oleic CCix:i) acids. were not obsen-ed at any of the stearic (Cls:o) acid
concentrations examined. Angelidaki and Ahring (1 995) reponed the same obsemation
.Accordin~ to the mass balance (Figure 7.3), acetate and stearic ( C I R : ~ ) acid accounted for
al1 the initially added stearic (Cis.o) acid up to day 20. Afier day 20, the decrease in the
rnass balance profile is Iikely due to degradation of acetate produced fiom P-oxidation of
Table 8.5: Free energy values for P-oxidation of iinoleate (CIX:~), oieate (CIS:~)
of stearic (CiXU) acid to paImitic (Cl6:~) acid becomes more feasible but in cornparison.
and stearate (C1g:o) to palmitate (CIG:O)
the conversion of linoleic (CIX:~) and oleic (Clg.1) acids remains more favourable
Reaction Stearate - 2 H - 0 + .4c' - Palmitate + 2H2 +- H- OIeate +- 2 H 2 0 + AC- -i Palmitate - + H
assumins the same LCFA concentrations.
AG": kl-mole-' 50.82 -27.8
8.2 Possible Pathways for Linoleic Acid Degradation
Several researchers proposed LCFAs to be P-oxidized by hydrogen producing
acetogens to acetate and hydrogen (Jeris and McCarty, 1965; Novak and Carlson, 1970;
Weng and feris, 1976). Hydrogenation of unsaturated LCFAs to stearic (Clg:~) acid prior
to 0-osidation has been proposed by Mackie et al. (1 991) and Novak and Carlson (1 970).
Linoleate - 2 H 2 0 -, Ac' + Pairnitate + H- I -106.38
Throughout this work, stearic (Cl*:-,) acid was not detected as an intermediate when
linoieic (Clg:2) and oleic (C18:l) acids were used as substrates. Several proposed
degradation pathways shown in Figure S. 1 for the conversion of C 18 LCFAs to palmitic
(C16:O) acid are based on data from this study. It is clear that complete LCFA
hydrogenation prior to B-oxidation is not necessary to produce the products observed.
In cultures fed stearic (C18:o) acid, a large amount of substrate remained undegraded
and no LCFA byproducts were observed suggesting that stearic (Clg:~) acid cannot
readil y undergo P-oxidation. Pathway 1 - 1 a- 1 b shows the conversion of Iinoleic
acid to stearic (CI8:O) acid is thermodynamically feasible however, the degradation of
stearic (CIS:O) to palmitic (&:O) acid is not possible under standard conditions. Based on
the calculated free energies, pathway 1-2-2a is not possible under standard conditions
since the conversion of oleic (Cls:~) acid to palmitoleic (C16:1) is thermodynarnically not
feasible. Because palmitoleic (C16:l) and paimitic (C16:0) acids were observed byproducts
from Iinoleic (Cig:~) acid, a Iikely route is 4-2a. To account for the oleic CI^:,) acid
observed during the conversion of linoleic (C1gZ2) acid, another possible pathway is 1-3.
No palmitoleic (Cl6:,) acid was observed when oleic (Clg:l) acid was used as a
substrate and a possibte route for its conversion to palmitic (&:O) acid is pathway 3 . The
overail free energy of conversion from iinoleic acid to palmitic (&:O) acid is
approximately -106.38 U-mole'' (Pathway 5 ) . It is unlikely a single enzyme can
accomplish this reaction in a single step. Pathway 5 is essentially composed of
hydrogenation, isomerization and P-oxidation steps and it is a combination of reactions in
pathways 1, 2, 3 or 4. Based on al1 the LCFA byproducts observed during the P-
osidation of linoleic (C18:2) acid, nvo pathways, 1-3 and 4-2a, likely mediated the
production of oleic (Clg,~), palmitoleic (C16:I) and palmitic (&:O) acids fiom linoleic
(CIR." acids.
At this stage, it is clear that the degradation mechanism is cornplex. The available
information in the literature does not elucidate the mechanism completeIy. Additional
work using radiolabeled tracers and pure cultures is required to thoroughly understand the
cornplete rnechanism.
palrnitoleic (C i6: 1) acid t
Pathway 1
Pathway 3
-78.57
palmitic (&O) acid
4 J Figure 8.1 : Proposed C 18 LCFAs degradation pathways. (Free energy values shown are in kJ-mole-' and were calculated assuming standard conditions. These complete reaction and fiee energy values are shown in Chapter 2)
9.0 INHIBITORY EFFECTS OF LLNOLEIC (Ci#& OLEIC (Cis:,) AND STEARiC (Cis:o) ACIDS ON ACIDOGENIC, ACETOGENIC AND METELANOGENIC ACTiVITY
(C18:0) acids showed rate decreases of 92%, 78% and 2S0A, respectively
The addition of 100 r n g ~ - l total LCFA (33.3 r n g . ~ " linoleic CI^, oleic (Clxl) and
stearic (Clx:~) acids, condition 6 in Table 9.2) lowered the butyrate degradation rate in
comparison to cultures receiving a single LCFA. A fùrther rate reduction was also
observed in cultures receiving 300 rng.~- ' total LCFA (100 rng-~- ' linoleic
oleic(C1~ 1) and stearic (Ci8:o) acids) in comparison to conditions 3 to 6. In contrast to
cultures receiving diethyl ether alone, almost complete inhibition (99% butyrate
degradation rate reduction) was observed in cultures receiving a total of 300 mg-^"
LCFA.
Table 9.2: Butyrate degradation rates for individual and mixed LCFA substrates
C- I
' ~ u ~ l i c a t e s . '~verage and standard deviation for triplkate samples, ?oral LCFA concentration = 100 rng-~-'. ' ~o ta l LCFA concentration = 300 r n g - ~ - '
Condition Exarnined
1. Control
Based on the Tukey's paired comparison procedure. the butyrate degradation rates for
Butyrate Degradation Rates ( p n - m n ~ ~ ~ - ' -da?.')
119.3'
each case esarnined \vas significantly different fiom every other case.
DiethyI ether had no significant effect on hydrogenotrophic methanogens (Figure 9.5).
The action of individual and mixed LCFA substrates on hydrogen uptake is shown in
Figure 9.6. The culture concentration used in this experiment had 1500 + 80 r n g - ~ - ' VSS.
Assuming first order kinetics (which showed r2 > 0.97 for al1 cases), the average
hydrogen uptake rate constants deterrnined for single and mixed LCFAs concentrations
are shown in Table 9.3. Based on the Tukey's paired comparison procedure (Box et al..
1978) at a 95 % confidence level. the differences in average rate constants for cultures
receiving 100 r n g - ~ - ' stearic (Clg,l) acid and those fed with linoleic (CIg:2) acid, oleic
Tinie, hrs Figure 9.5: Duplicate conuol cultures hydrogen profiles with and without diethyl ether. (DE = diethyl ether, Average for duplicate results shown)
Figure 9.6: Hydrogen profiles for cultures receiving individuai and mixed LCFAs. (SA = steanc ( C l g : ~ ) acid, OA = oleic (Cls:!) acid, LA = linoleic acid; Duplicate controls; Averages for triplicate samples, error bars represent standard deviation for the samples)
(C 1 g ,) acid and mixtures of three LCFAs (Condition 6 and 7 in Table 9.3) were
statistically significant. There was no evidence of any synergestic interaction due to
mixtures of LCFAs on hydrogen consumption.
9.2 Discussion of Results
9.2.1 Glucose degradation
Table 9.3: First order rate constants for hydrogen rernoval in the presence of individual and mixed LCFAs
Linoleic (ClS.:) acid was more inhibitory to glucose degradation in comparison to
oleic (Cis i ) and stearic (Cig:o) acids. Based on the degradation rates, inhibition caused by
stearic ( C i ô , ~ ) and oleic (CIg:~) acids was approximately the same. No comparison has
been made in previous research of the effects of single saturated and unsaturated LCFAs
on glucose degradation. In comparison to the rate observed for stearic ( C I R : ~ ) acid, the
Condition Examined
1 . Control 2. Control -+ DE 3. 1 O0 mg^" S A
presence of a single double bond had no major effect on glucose degradation.
addition of a second double bond (Iinoleic (C'S:~) acid) decreased the rate sign
J
First Order Rate Constant, d-'
0.409' 0.388'
0.374 0.006'
However,
ificantly.
4. 100 r n g - ~ " OA 5. 100 mg-^" LA 6. 33.3 r n g - ~ " LAIOAISA~ 7 100 mg^" LA/O.WSA"
0.260 I 0.006~ 0.263 I 0.013' 0.245 ,+ 0.01 1' 0.255 +. 0.003~
'~u~l ica tes , 'AveraSe and standard deviation for triplicate samples ' ~ o t a l LCFA = 100 r n g ~ ' ' . " ~ o t a l LCFA = 300 rng.L-', LA = linoleic (C 1s.:) acid, 0-4 = oleic (C1g:I) acid, SA = stearic (Cis:~) acid.
One possible explmation may be related to the effects of LCFAs on ce11 membrane
receptors that are responsible for glucose uptake. Linoleic (C18:2) acid may have a higher
binding affinity for these receptors in comparison to oieic (C18:l) and stearic (C18:0) acids
hence causing a greater inhibitory effect on glucose uptake.
The combined effect of mixed LCFAs (Conditions 7 in Table 9.1) on glucose
degradation shows that there is no synergistic interaction to hnher lower the rates below
those observed for linoleic (Cis:?) acid (Condition 5 in Table 9.1). However, the
concentration dependence of linoleic (Ci8:2) acid inhibition on glucose degradation is
clearly shown.
The effect of mixed LCFAs on glucose degradation was examined by Hanaki et al.
( 1 98 1 ) using 2500 mg^" VSS at 37 OC. They reponed glucose fermentation was
uninhibited by increasing concentrations (O to 2000 mg-^-' as oleate) of a fatty acids
mixture containing saturated Cie to C18 LCFAs and mono-unsaturated LCFAs fiom Ci? to
CIg. The difference between Our work and that reported by Hanaki et al. (198 1) is that
the parameters used to rneasure glucose degradation are not the same. Our work reported
direct measurement of glucose using an enzyme assay while Hanaki el al. ( 1 98 1) reported
acetate (Cz ) , a byproduct of glucose degradation as their measure of inhibition. The error
reported in their work is that in addition to acetate (C2) decived fiom glucose degradation,
they also measured acetate (C2) arising from LCFA P-oxidation. As a result, the data
reported by Hanaki er al. (198 1 ) is an inaccurate measure of the inhibitory effects of the
fatty acid mixture on glucose fermentation. Additionally, in comparison to the effect of
LCFAs mixtures reported by Hanaki et al. (1 98 l), Our work examined the effect of
individual and mixed C 18 LCFAs on glucose degradation.
9.3.2 Butyrate Fermentation
Diethyl ether inhibited butyrate fermentation in comparison to cultures without diethyl
ether. Single and mixed C 18 LCFAs affected butyrate degradation at al1 the conditions
examined. At 100 mg-^" LCFA, butyrate degradation rates decreased as the number of
doubIe bonds increased in the Cl8 homologous series from stearic (Clx:o) acid to linoleic
( C l 4 acid..
Based on the degradation rates, cultures receiving mixed LCFAs were inhibited more
in comparison to those receiving individual C 18 LCFAs. This synergistic relationship
was more obvious as the LCFA mixture concentration increased. This synergism may
have been caused by inactivation of receptors responsible for transpon of butyrate into
cells. If a specific LCFA such as linoleic (CIR:~) acid is able to bind to a butyrate
receptor. its action may enhance the binding of oleic acid and / or s t eak (Clg:~)
acid to other butyrate receptors. Synergistic effects of LCFAs on anaerobic organisms
have been previously reported by Canovas-Diaz et al. (1 99 1 ) and Koster and Cramer
( 1 987). Canovas-Diaz et al. ( 199 1) reported that decreased degradation rates were
observed when oleic and myristic (Ci+o) acids were added together as a mixture,
in comparison to cultures degrading the individual acids. In the presence of capric (Clo.o)
and myristic (Cl4:o) acids, Koster and Cramer (1987) reported lauric (Ci2:o) acid to have a
synergistic inhibitory efîect on rnethanogenesis.
Angelidaki and Ahring (1992) reported that higher than 500 rng*~-' oleic (Cltt,i) acid
in hibited but yrate and propionate fermentation. Although a threshoid value causing
inhibition was not determined, these studies showed that 100 mg^" oleic (CI~: l ) acid
inhibited butyrate fermentation. Differences between our resuIts at 21 OC and those
reponed by Angelidaici and Ahring (1 992) at 55 OC may be a reflection of temperature
variation but also to culture adaptation. It should be noted that in their report (Angelidaki
and Ahring, 2993) the VSS concentration was not available and this may also contribute
to the difference between the two studies,
9.2.3 Hydrogen Consumption
Cunently no research has reported on inhibition of hydrogenotrophic methanogenic
by direct measurement of hydrogen for single LCFA compounds and LCFA mixtures at
several concentrations. Research by Demeyer and Henderickx (1967) has, however,
reported that linolenic (C18:3) acid inhibits methane production fiom H2 and COz and
Hanaki rf al. (198 1 ) reponed inhibition of hydrogenotrophic methanogens by LCFAs.
In the presence of a LCFA mixture at 3 7 O C , Hanaki et al. (1 98 1 ) found
methanogenesis to be inhibited although al1 o f the hydrogen added was readily utilized.
Similarly, in our studies with LCFA mixtures, even though hydrogenotrophic
methanogenesis was slightly inhibited, the hydrogen was consumed to undetectable
levels within approximately 8 hours, similar to control cultures (Figure 9.5).
Stearic (Cix:a) acid did not affect hydrogen consumption in comparison to control
cultures with diethyl ether. The lower first order kinetic constants observed for linoleic
(C18:2) and oleic (Cii:]) acids, in comparison to stearic (Cis:o) acid, illustrate the inhibitory
effect caused by the presence of double bonds on hydrogen consumption. Although
lower first order constants were obsemed for oleic (Cis:]) and linoleic (Clgz) acids, the
hydrogen consumption was unaffected by addition of a second double bond.
First order rate constants, determined for cultures receiving 100 and 300 mg-L" mixed
LCFAs and unsaturôted LCFAs, were not significantly different based on the Tukey's
paired corn parison procedure. These results suggest that there is no inhibitory synergy
between LCFA mixtures on hydrogen consumption by hydrogenotrophic methanogens.
Mixed LCF.4s ranging fiom 100 to 300 mg-^-' did not inhibit hydrogen consumption
slightly. Assuming negligible effects of stearic (C18:0) acid on the first order rate
constant. the inhibitos. effects of linoleic ( C I S : ~ ) and oleic acids in LCFAs
mixtures are approximately the sarne as cultures receiving only linoleic ( C 1 4 acid or
oleic (Clg:~) acid.
9.3 Surnmary
Glucose fermenters were affected under al1 the conditions examined. The effect of
LCF.4s on glucose consumption is as follows: 100 mg-^-' linoleic (CIsc2) acid z 300
r n g - ~ " total LCFA > 100 mg-L" total LCFA > 100 mg-L-' oleic (C18,i) z 100 rng-L-'
stearic ( C , S . ~ ) acid. For butyrate consumption the inhibitory effect of 300 r n g - ~ - ' total
LCFA > 100 rng.L-' total LCFA > 100 mg^-' iinoleic (ClsZ2) > 100 mg^-' oleic ( C l s , ~ ) >
100 rng-L-' stearic (Cip:i) acid Although inhibition of hydrogenotrophic methanogenesis
was statistically sigificant between cultures receiving stearic (C18:o) acid versus those
receiving linoleic (Ci8:2) acid, oleic (C18:l) acid and LCFA mixtures, the inhibition was
minimal as most of the hydrogen was consumed within less than 10 hours.
10.0 DISCUSSION OF METHANOGENIC STUDIES AND LCFA LNHIBITION MECaANISM
10.1 Methanogenic Studies
10.1.1 Aceticlastic Methanogens
Under anaerobic conditions and in the absence of inhibitors, acetate conversion to
methane and carbon dioxide by aceticlastic methanogens is unaffected. However, in the
presence of LCFAs, aceticlastic methanogens are inhibited. During the degradations
studies, Iinoleic ( C 4 , oleic (CIS:I) and stearic (C 1g:o) acids at dl concentrations
exarnined, inhibited aceticlastic methanogens.
Variation in acetate profiles during the degradation studies for the three C 18 LCFAs
esarnined rnay be due to diflerent concentrations of LCFA byproducts produced. For
esample, larger amounts of palmitic (CtG:O) and myristic (Cta:o) acids were produced
during P-oxidation of 100 mg^-' linoleic acid compared to the amounts produced
during degradation of 100 m g - ~ - I oleic acid. These compounds may inhibit
aceticlast ic methanogens and affect acetate consumption.
The binding of C 18 LCF.4s to ceII receptors responsible for transport of acetate into
the rnethanogenic cell may play a role in inhibition. This possibility may explain why
acetate accumulated during LCFA degradation studies but was often consumed during
the inhibition studies. Assume that LCFAs and acetate molecuIes are able to bind to
these receptors sirnultaneously. Aiso assume that acetate when bound is transponed into
the ce11 while LCFAs when bound cause irreversible damage. During the LCFA
degradation studies, no acetate was initially available to compete with LCFAs for acetate
receptor sites. Therefore, most of the receptors may have been damaged so that when
acetate was produced, it accumulated in solution. As time progressed, more receptors
became available and the amount of acetate decreased as those aceticlastic methanogens
that were able to survive increased their population size.
During the inhibition studies, acetate was added simultaneously with linoleic (C18.z),
oleic ( C l 8 , ) or stearic CI*:^) acids. If binding is competitive, at low LCFA
concentrations, acetate concentration may have been sufficient to bind to most of the
receptors. This allowed acetate consumption to proceed. However, as LCFA
concentrations increased, LCFA molecules out-cornpeted acetate for receptor sites. The
higher the LCFA concentration, the less competitive acetate became and the acetate
consumption rate decreased.
In cornparison to cultures receiving Iinoleic or oleic (Cig:l) acids, acetate was
consumed initially and no lag was observed for cultures receiving stearic (C18:o) acid.
Stearic (Clgo) acid may behave differently to linoIeic (C 18~2) or oleic CI^:^) acids by
binding reversibly to the ce11 receptors without causing any damage. Hence, acetate
binds and is transported into the ce11 when stearic (Clg:~) acid is released from the
receptor.
That less aceticlastic methanogenic inhibition is caused by saturated LCFAs is
supported by Demeyer and Hendrickx (1967) who reported that LCFA inhibition
increased with increasing the number of double bonds. They aIso demonstrated that
LCF.4 inhibition increased as the LCFA carbon chain increased.
10.1.2 Hydrogenotrophic Methanogens
ln this study, hydrogenotrophic methanogens were inhibited to a much lesser extent
t han acet iclastic rnethanogens. Hanaki et al., (1 98 1 ) reported hydrogenotrophic
methanogens were less sensitive to a LCFA mixture in comparison to aceticlastic
methanogens. Although some inhibition was observed, no clear trends can be
determined. Under al1 the LCFA conditions examined, inhibition of hydrogenotrophic
methanogens is not expected to impact anaerobic degradation to the extent that inhibition
of aceticlastic methanogens will.
10.2 LCFA Inhibition Mechanism
The inhibitory effects of LCFAs on anaerobic fermentation have been exarnined
previously (Rinzema et al., 1994; Angelidaki and Ahring, 1992; Koster and Cramer,
1987; Hanaki et al., 198 1 ; Demeyer and Henderickx, 1967). However, the molecular
basis of LCFA inhibitory effects on anaerobic organisms or more specifically, on
aceticlastic and hydrogenotrophic methanogens has not been examined. Figure 10.1
shows several possible pathways for LCFAs ceIlular uptake. These pathways are based
on LCFA literature research in anaerobic degradation, food presewation and uptake into
eukaryotic cells. It is possible LCFAs may follow either of three routes: activate the
outer membrane sensory proteins and be transported into the cell, inactivate the outer
membrane enzyme or integrate into the cellular membrane.
Researchers have proposed a glucose uptake mechanism for eukaryotic cells (Van
Winkle, 1995; MueckIer et al., 1985) and ~ising the glucose model, a similar approach
can be developed for LCFA uptake into bacterial cells. LCFAs may first activate a
sensory membrane protein which causes the ce11 to initiate the uptake process. Mangroo
er al. ( 1 995) and Nunn (1986) have examined the transport mechanism of LCFAs into
adipocytes and Escherichia coli, respectively. Both researchers suggest a transponer
L-_-* Disrupt proton cliemical potential
IXFAs b Activaie .-+ I'ransportcd b B-oxidation - no inhibition but
Inactivate outer Disnipiion o f membrane function for membrane example, transporter protein enzyriies
out er irito the ceIl iiicnibraiie enzy riies
L lntegation into the cellular Phospholipid formation meiiibraiie
nossible bvproduct inliibiiiori.
b Iiiact haie enzymes
t- Disniption o f membrane function for example, transporter protein
hisertion into the membrane causes cell lysis
Figure 10.1 : Possible IdCFA inhibition pathwiiy s
protein, fadL. is responsible for the transmembrane movement of LCFAs into cells. But
the mechanism is not completely understood at the molecular level. It is possible that the
fadL protein is synthesized and inserted across the ce11 membrane afier outer membrane
sensory proteins detect the presence on LFCAs on the ce11 membrane. Subsequently, the
fadL protein is inserted into the ce11 membrane to initiate the LCFA transport process.
When transported into cells, LCFAs may undergo P-oxidation (Weng and Jeris, 1976;
Jeris and McCarty, l96j), bind to cellular enzymes (Ferdinandus and Clark, 1969) or
disrupt the proton chemical potential (Sikkema et al., 1995). During P-oxidation,
hydrogen producing acetogens degrade LCFAs to shorter chain LCFAs, acetate and
hydrozen. LCFAs can also bind and inactivate metabolic enzymes. Octanoic (CR) acid
was reported to cause inhibition of several enzymes responsible for the synthesis of new
LCFAs from precursors such as pyruvate, in lipogenesis and intermediates of the
tricarbosylic acid cycle (Ferdinandus and Clark, 1969). LCFAs may disrupt the proton
chemical potential across cellular membranes and inactivate reactions such as ATP
synthesis. Obstructing membrane transport tunction such as interaction with ATPase,
LCFAs are also able to uncouple the proton potential used for oxidative phosphorylation
(Sikkema er al., 1995). Rottenberg and Kashimoto (1986) and Fay and Farias (1977)
reported simulating proton and potassium ions leakage across various membranes upon
addition of LCFAs.
Another cellular pathway for LCFAs is integration into membranes (Figure 10.1 ).
When present in membranes, LCFAs may undergo transformation to become
phospholipids, disrupt membrane fùnctions or cause cell lysis. Phospholipids are a major
component that is essential to the function of cellular membranes. Greenway and Dyke
( 1 979) repoxted approximately 0.9 % linoleic (C 1 8 : ~ ) acid was incorporated into the
p hospholipid component of the outer membrane of Staphylococcus aureus and free
linoleic (Clg::) acid seemed to be the growth inhibitory substance and not the linoleic
( C 1 8 . 2 ) acid component present in the membrane lipid bilayer.
One role of proteins in cells is for transpon of molecuies into or out of cells.
Membrane proteins responsible for this function are called transporter proteins. Possible
damage caused to these proteins by C 18 LCFA was discussed previously.
LCFAs are able to insert into cellular membranes and cause ce11 lysis. Some research
has been conducted on the effect of LCFAs on ce11 lysis. Using Staphylococclts airrerts
Greenway and Dyke (1 979) reponed the release of a 260-nm material was dependent on
the linoleic acid concentration. They aiso used stearic (Ci8:o) acid at an equivalent
linoleic ( C ~ S . ~ ) concentration greater than 50 r n g ~ - ' and reponed that the 260-nm
substance was not released.
Linoleic (Cis::) acid is reported to behave as a surfactant by altering the interfacial
tension between the bacterial membrane and the bulk aqueous phase of the growth
medium (Greenway and Dyke, 1979). The ability for a compound such as linoleic ( C i 4
acid to migrate to cellular surfaces and lower the interfacial tension (Le. increase the
wettability of the surface) is related to its surface tension and sohbility properties. For
esample, in comparison to linoleic (C18:2) acid, stearic (C18:o) acid was reported not to
inhibit gro~kth at 30 "C since it is a much poorer surfactant (Greenway and Dyke, 1979).
Thus. LCFAs such as linoleic acid may be more easily degraded and yet cause
more inhibition in comparison to stearic (Crg:~) acid.
Stearic (Clg:~) acid was less inhibitory to aceticlastic methaogens in cornparison to
linoleic (Clrc:?) and oleic (C 18~1) acids. The soap solubilities of the Ci* LCFAs
homologous series for linoleic (Clg:~) and oleic (Cls:~) acids are approximately equal
(Irani and Callis, 1960) and the same solubility is assumed for stearic (Clg:~) acid.
Therefore, a factor other than the solubility that is likely affecting the inhibition, is the
chemical structure. Stearic ( C l s : ~ ) acid has no double bonds and the molecule is less rigid
cornpared to oleic ( C I R . ~ ) and linoleic (C18:?) acids. It is possible that this structural
difference may cause ce11 membrane receptors to have a low binding affrnity for stearic
(&:O) acid and thus, affect its degradation and inhibitory characteristics.
11.0 SLMMARY AND CONCLUSIONS
11.1 Summary
1 1 .l. 1 LCF A Degradation
Linoleic (CI*:-) and oleic (CIg:l) acids were degradable, however, in cornparison,
stearic acid was not completely B-oxidized at al1 the concentrations exarnined (O to
100 mg-~- l ) . Palmitoleic (C16:1) (less than 10.5 r n g ~ - ' ) , hexanoic (Cs) (less than 37.0
rng-~") and trace arnounts of lauric (Ciz:0) acids were observed in cultures receiving 100
r n g - ~ ' ' linoleic (Cis.:) acid. Oleic (CI~ : l ) , palmitic (C16:0), myristic (Cli:o) and acetic acids
were detected in al1 cultures receiving linoleic ( C 1 4 acid. Within less than 35 days,
linoleic (Cis:z) acid was degraded at al1 concentrations examined. The length of time to
comp~etely remove palmitic (Cl6:*) and myristic (Cld,~) acids from the system was
concentration dependent with long rernoval times observed at high dosages of linoleic
(Cin:2) acid. In cultures receiving higher than 10 r n g - ~ - l linoleic (Clg,2) acid, acetate (Cz)
accumulated from P-oxidation of LCFA byproducts.
Palmitic (Cll,o). myristic and acetic (Cz) acids were observed at al1 oleic (Cla:,)
acid concentrations examined. In cornparison to linoleic (C1s:2) acid, palmitic (C16:0)
and myristic (C li,o) acids were removed within shorter time periods in cultures receiving
oleic (C1s,l) acid. This suggests that oleic (Cia:l) acid might be less inhibitory than
linoleic (Ci8:2) acid to organisms responsible for B-oxidation of palmitic (C16:0) and
myristic (Cli:o) acids. Although acetate (C:) was consumed at al1 oleic (Cig:l) acid
concentrations examined, inhibition of aceticlastic methanogenic was observed in
cultures receiving higher than 50 r n g ~ ~ l .
No LCFA P-oxidation byproducts were observed in cuItures fed with stearic
acid. However, acetate (C2) was produced at ail concentrations examined. About 50 to
60 % of the stearic (C18:o) acid added remained undegraded after day 5 5 . Acetate ( C l )
accumulated to between 30 to 50 mg^" in cultures receiving greater than 30 r n g ~ . '
stearic (Cis:o) acid. It is unclear why stearic ( C i 8 : ~ ) acid is not easily degradable in
comparison to linoleic (C18:2) and oleic (Cig:l) acids. However, as previously discussed
in section 8.1, it is Iikely that the conversion of linoleic ( C I ~ ? ) and okic (Cig:~) acids is
favored thermodynamicaIIy. Based on reaction free energies, the most feasible
degradation pathway for linoleic acid is conversion to paimitoieic (C16:1) acid and
then hydrogenation to palmitic C CI^:^) acid. For oleic (Crs:~) acid, direct conversion to
palmitic (Clb .* ) acid is thermodynamically the most likely pathway.
1 1.1.2 LCFA Inhibition
-4. -4 cidoger~s
Glucose degradation by acidogens was affected at al1 individual and mixed LCFA
concentrations esamined in comparison to control cultures. Linoleic (CIS:~ ) acid by itself
caused the greatest inhibition in comparison to cultures receiving oleic (Cis:~) or stearic
'l'atilc A.3: Estiniatioii o f coristants for Aiitoiric's cquat ion Vapor I'rcssure, nini Hg Compound CHj(CH2).lC001i Cl-13(CH2)r,COOH CIH3(CH2)nCOOH CHj(CH2)loCOOH CH1(CH2)l 2COOH CHJ(CH~),&OOH CCIx(CH2)16COOH CHj(CH2)~CH=CH(CH2)7COOH "' CH~(CH2)~CH=CHCH2CH=CH(CH2)1C001-I "' CH~(CH~)ICH=CH(CH~)~COOH ("
Antoine's equation for vapor pressure log 10 P = -0.05223(a/T) + b """~ssuine same constants a and b as for stcaric (Ciw 0) acid (CI-~,(CHI)I(,COOH) (no data available). '"~ssume same constants a and b as for palniitic (C16:iJ acid (CH3(CI-12)irCOOH) (no data available). '4"5' Values for a and b were estimated frorn solving Antoine's equation under the two given teinpcrature
"' Calculated from Antoiiie's equat ion for vapor pressure (sec Table A. 3) log 10 P = -0.OS223(a/T) + b (Dean, 1999)
'2' Ralston and Hoerr ( 1943) "' Henry's constant = vapor pressure (atm) 1 solubility (molesi.'). '""" Assume solubility the samc as stcariç (Cl n:o) acid (CH3(CH2),(,COOH). '" Assume solubility the sanie as palinitic (Clc:ii) acid (CI.li(CH2)l.lCOOH).
Compound Vapor pressure, atm " )
Solubility (water) L - 1 (2 )
Solubil it y, niolesl"
Henry's Constant a t iw~mole~ ' (')
Tiiblc A.5: Estiiiiiiiioii of ioiiizcd LCFA frcc ciicrgv of foriiiiiiioii
H = Henry's constani, valiics iakcn froiii Tiiblc A.4. ('' D G ~ ( 1099) "'K. for CC, Io Clfi wcrc csiiiintcd froni il linciir ~Wi ipo l i i t i ~ i i of K. niliics for C: Io C,, iicids. K ii VII~ICS for Cl(, iiiid Cl ,' iirc ~ISSIIIIIC~ IO bc flic siiiiic.
"' Ka valiics for CI* LCFAs iirc iissiiiiicd io bc ilic sariic. '" CCrclciila~cd viiliics iirc bciwccii 7 io IO %i grciiîcr t l i i i i i viiliics rcpricd by (Thiiiicr O( trl. , 1077).