IV. Plants as Remediation Structure for Organics

PHYTOREMEDIATION OF SOILS



73



where K , is the Henry’s Law constant, [A(aq)J is the concentration of the

component in the solution phase, and PA is the partial pressure of A in the gas

phase. Ryan et ul. (1988) compiled a list of Henry’s constants for priority

pollutants. Some attempts have been made to define a value for the Henry’s

constant above which volatilization is an important mechanism for bioavailability, but one must consider other chemical properties such as adsorption

to the soil, microbial decomposition, and solubility in water. For example, the

polyaromatic hydrocarbons (PAHs) are generally considered semivolatile. However, the microbial decomposition of the lower molecular weight compounds

(such as naphthalene) and/or strong adsorption to soil components for higher

molecular weight PAHs (such as pyrene and benzralanthracene) make volatilization from soil a minor pathway for this class of compounds.



2. Water Phase

Lipophilicity is the most important property of a chemical in determining its

movement into and within a plant. Lipophilicity is the “balance between the

affinity of the chemical for aqueous phases and that for lipid-like phases” (Bromilow and Chamberlain, 1995). This property determines the ease of movement

across plant membranes. Lipophilicity is related to the partition coefficient of the

pollutant between I-octanol and water (Kow).The K , , is one of the most widely

available experimental parameters for xenobiotics. The values of K O , cover the

wide range from 10-2 to 10’0. For ease of discussion these numbers are often

reported in log form. A small KO, is indicative of high water solubility and low

lipophilicity, and high values are associated with compounds that have high

lipophilicity and low water solubility. Organic contaminants in soil can transfer

to roots in the soil-pore water by diffusion and by mass flow. In either case, the

transfer is mediated by the aqueous phase. The spot-to-spot variations of soil

texture, organic matter composition, water conditions, and distribution of organic contaminants make the root exposure to soil organic contaminants a highly

heterogeneous and constantly changing phenomenon. This explains why many

experiments have been conducted in hydroponic cultures in which an idealized

condition can be easily created and maintained. For a hydroponic system, the

concentration of the test compound can be kept constant to all parts of the root

during the entire experimental period.

Compounds that are most water soluble will be most available for mass transport and diffusion into the rhizosphere. However, very soluble compounds with

little afYinity for the soil solids also will be subject to leaching out of the root zone

and can become physically unavailable for root uptake. Marginally soluble components will not move with the water and require the root to grow near them

before uptake is possible. Positively charged organics will tend to be retained by

the soil’s cation exchange sites and their availability for assimilation will be

reduced. Negatively charged compounds are excluded from the generally nega-



74



S. D. C U " G H A M



ET AL.



tively charged soil surfaces, making them more available to the roots but further

subject to leaching.

Research on the plant uptake of organic chemicals from soil has centered on

pesticides and a few high-profile contaminants (such as PCBs and dioxin). Pesticides have been emphasized because of the importance of plant uptake in their

function and their potential for effects on the food chain and nontarget crops.

Also, knowledge of a pesticide's environmental fate is necessary for its registration and sale. Plant uptake models have been developed using experimental data

for those compounds that have been studied and these models allow us to predict

plant uptake of organic compounds for which no experimental data exist.



3. Solid Phase

Conceivably, an organic compound adsorbed to a soil particle can directly

transfer into a plant root via solid-to-solid partitioning. This route of transfer may

allow us to rationalize some putative high root uptakes of very lipophilic compounds with high vapor pressures as well (McMullin, 1993). This concept is

technically feasible and well demonstrated in dermal sorption studies in mammalian systems; however, due to the vast surface area in a soil and the relatively

low surface area of a root in comparison, it is hard to imagine this as a viable

remediation strategy. Nevertheless, we do not wish to rule this out as either a

mechanism of plant uptake or a potential remediation technology in some creative form.



B. XENOBIOTIC

FATEIN A PLANTSYSTEM

1. Transport within the Plant

Physiological factors of plant roots that control uptake of organic chemicals

have been summarized by McFarlane (1995). Briefly, water and dissolved constituents can move easily and relatively unimpeded from the soil solution into the

root's apparent free space, that area of the root outside the endodermis and the

Casparian strip. The apparent free space is characterized by cortex cells with

porous cell walls and many voids. The Casparian strip in the endodermis is a

waxy barrier that inhibits movement of water into the interior of the root. At this

point, all water, solutes, and non-aqueous phase liquids must pass through cell

membrane at least two times. For most xenobiotics (both ionic and nonionic

substances), this movement a p p r s to be a passive process depending upon

retention by the membrane, solubility in water, and diffusion. Many observations

on xenobiotic root uptake and root-to-shoot translocation have been made (Devine and Vanden Borden, 1991). However, most were aimed at describing the



PHYTOREMEDIATION OF SOILS



75



behavior of specific compounds of interest. Here we only review those studies

that are instrumental in revealing the underlying general principles governing the

processes.

Movement of nonionized organic chemicals into the root consists of the equilibration of the solution phase outside the root with the solution in the apparent

free space and sorption of the compounds onto the root surface. Shone and Wood

(1974) examined the uptake of the herbicide sirnazine by barley growing in

solution culture. One of the more interesting observations of their research was

that the xylem translocation of simazine apparently was restricted because the

concentration of simazine in the transpiration stream was less than the concentration in solution. A transpiration stream concentration factor (TSCF) was developed to describe this behavior.

TSCF



=



concentration in xylem sap

concentration in external solution



In the case of simazine or any other compound whose xylem translocation is

apparently restricted, TSCF is less than 1. 0. Other triazines also were found to

have TSCF less than 1.0 (Shone el a!., 1974). Shone and Wood (1974) also

defined a root concentration factor (RCF).

RCF



=



root concentration

external solution concentration



This line of study was extended by Briggs et al. (1982) to a series of other

pesticides (0-methylcarbamoyloximes and substituted phenylureas). They found

no relationship between RCF and TSCF and suggested that root accumulation of

one class (triazines) was mostly physical adsorption to the surfaces of the roots.

For other classes (carbamoxyloxime and phenylureas), an empirical model was

made to relate root uptake by intact barley plants to lipophilicity:

log (RCF - 0.82)



=



0.77 log KO, - 1.52.



This equation is expected to vary somewhat among plant species depending upon

the composition of lipids in the roots.

They found that the TSCFs were less than 1 .O for all 18 compounds studied,

and that TSCFs showed a bell-shaped relationship with log K , , values.

TSCF



=



0.784 exp



[(log KO, - 1.78)]*

-2.44



The maximum TSCF was observed for log KO, of about 1. 8. They also plotted

those literature data with sufficient details and found that despite the diversity of

plant species, compounds, and experimental techniques, they largely conform to

the same relationships they found between log K , , and RCF and TSCF. The

general conclusions of Briggs et al. (1982) were further validated by Hsu et al.



76



S. D. CUNNINGHAM ET AL.



(1990), albeit with a different maxima, on a series of herbicide (cinmethylin)

analogs using a root pressure chamber technique.

While the linear relationship between RCF and log KO, is immediately intuitive, the log KO, and TSCF relationship requires explanation. It seems to be

intimately related to the pathways from root surface to xylem vessels situated in

the stele. The root pathways for water and solutes have received reviews in the

past (Weatherley, 1975). The salient feature of the main pathway is that a molecule has to cross cell membranes at least twice, once before the Casparian strip

(CS) barrier on the endodermis, and once after that, to gain access to the xylem

in the stele (Weatherley, 1975; Clarkson and Robards, 1975). This is referred to

as the symplastic pathway. The apoplastic pathway, available only near the tip of

the root where the CS has not yet developed well, usually accounts for an

insignificant amount of the total flux unless the root is disturbed (Moon et af.,

1986) or stressed (Hanson et al., 1985; Skinner and Radin, 1994). Compounds

with low log KO, values can move through the intercellular space along with the

mass flow of water until reaching the CS barrier. There, the two membrane

crossing steps are slow due to low lipophilicity. On the other hand, delivery

efficiency of compounds with higher log KO, values through the pre-CS barrier

parts is low due to low water solubility and losses to lipophilic tissue constituents. But the rate of their membrane crossing is more rapid, compensating for the

earlier slower passage.

Once compounds are partitioned into lipophilic membranes interior to the CS

barrier, they need to be desorbed into aqueous solution in order to go into xylem

vessels. Conceptually, the entire root pathway may be simplified as a step of

partitioning into the post-CS membrane and a desorption step off this membrane.

The first step of aqueous-to-lipid partitioning obviously favors more lipophilic

compounds. However, the second lipid-to-aqueous desorption step favors more

hydrophilic compounds. The interplay of these two crucial rate-limiting steps

would produce the observed relationship between log KO, and TSCF (Briggs et

al., 1982; Hsu et al., 1990).

In soil the idealized hydroponic condition is compromised. Sorption of organics to soils can limit their availability to roots. For many organic compounds, one

can assume that the adsorption is a linear function of concentration and organic

carbon content of the soil,

9e = K d f ,



c,



wheref,, is the fraction of organic carbon in the soil, C , is the concentration of

the organic compound in the soil solution, and Kd is the linear adsorption coefficient. This equation suggests that as organic carbon content increases, adsorption

increases and availability for root uptake decreases. It is estimated that in soil the

log KO, for maximum TSCF shifts down by about two units (Hsu et al., 1990).

In the field of phytoremediation, not much attention has yet been paid to the

mechanistic aspects of root uptake and xylem translocation. Most studies use



PHYTOREMEDIATION OF SOILS



77



endpoint analyses of shoot contents of radiolabeled test compounds. Such studies

are always complicated by shoot metabolism and severely limited by the availability of radiolabeled compounds. Simple techniques are now available to allow

for mechanistic experimentation with nonradiolabeled compounds free of shoot

metabolism complications. This kind of mechanistic approach can yield crucial

information to help design the most optimal phytoextraction scheme for important soil organic contaminants.

Conceptually, the total amount of organic compound phytoextracted into the

(easily harvestable) above ground fraction can be defined as:

Amount delivered to shoot = Conc. in sap



X



Volume of sap/time unit.



The xylem sap can be obtained from the decapitated stem near the soil line

either by applying pressure to the root of a potted plant in a modified pressure

chamber as that used in Hsu et al. (1990) or by applying negative pressure to the

cut stem (Gil de Carrasco et al., 1994; Ambler et al., 1992). Two cautions need

to be observed to obtain xylem saps matching those of intact plants: ( I ) adjust the

sap expression rate to match that of the sap volume flow in intact plants, and (2)

avoid the initial sap sample which may contain artifacts associated with a given

technique (Else et al., 1994). Due to the generally simple composition of xylem

saps, the presence of a target organic compound and its metabolites can be

readily quantified by analytical techniques. Here the analytical challenge mentioned earlier (Section 1I.A) is quite easily met. This is evidenced by the successful quantification of low concentrations of the natural plant growth regulators

abscisic acid and zeatin in xylem sap (Ambler et al., 1992; Davies and Zhang,

1991). The whole-plant xylem sap flux rate can be measured with a stem-flow

gauge. The accuracy of the xylem mass flow measured with this method has been

validated in many studies (Baker and Van Bavel, 1987; Steinberg et al., 1989;

Dugas, 1990; Devitt et ul., 1993).

The adaptability of the xylem sap expression techniques and the xylem mass

flow measurement technique makes them particularly suitable for phytoextraction research. Both techniques can be used for both laboratory and field experiments with nonradiolabeled compounds. For laboratory studies, different plants

can be grown in soil with the target organic contaminant. The sap flow can be

measured, and xylem sap obtained for analytical quantifications. Different xylem

sap concentration values from different plants multiplied by their respective total

sap volume flow rates will generate the total amounts extracted by test species.

The study can be run at different times of day to cover diurnal variations of xylem

flow rates and sap contaminant concentrations. The method can also detect

whether the build-up of the target compound in shoot causes any self-limitation

of further phytoextraction by an inhibition of sap flow rate. By using these

monitoring techniques, different innovative methods can be tested to see if they

produce any improvements on net phytoextraction. For field studies, sap analysis

and sap flow measurements can be made to existing plants in a contaminated site.



78



S . D. CUNNINGHAM ET AL.



These measurements can quickly lead to the identification of the most advantageous plant species for phytoremediation.

Although this chapter deals specifically with organic compounds, the same

xylem sap analysis and sap flow measurement can be effectively applied to the

mechanistic study of phytoextraction of other soil contaminants, particularly

inorganics. Knowing their xylem sap contaminant content as well as the chemical identity of the chelating compound(s) would greatly help us in pinpointing the

potential rate-limiting step in phytoextraction.

Despite these general rules, there are distinct variations among plants in their

ability to accumulate organics into the roots. Selecting plant species and varieties

for maximizing this trait, however, has not been approached systematically. The

subject has been studied with certain pesticides, but few data exist for other

organic pollutants. The best-studied cases seem to involve chlorinated organic

insecticides. For example, beets, turnips, potatoes, and radishes tend to accumulate less than do carrots (Lichtenstein and Schulz, 1965). Sugar beet roots accumulated more dieldrin than carrots, potatoes, corn, and alfalfa (Harris and Sans,

1967). Different varieties of carrot can have four-fold differences in endrin uptake (Hermanson et af., 1970). More detailed comments on the influence of plant

properties on root uptake of organic compounds are provided by Bell (1992) and

Shimp and co-workers (1993).

In addition to root uptake, the general microbial stimulation in the root zone

suggests that the most logical starting point in selecting species is to focus on the

plant root. Species with extensive and fine root systems should have the greatest

potential for enhancing bioremediation. These roots and their associated microflora would be more apt to have greater soil/root surface contact and be able to

penetrate small pores than species with a coarse taproot system. Mycorrhizae

may provide additional advantages because of their fine architecture ability to

increase the effective surface area of the root as well as their microbial metabolic

traits.



2. Metabolism within the Plant

Humankind was not the first to create biologically disruptive organic compounds and place them in a soil-plant environment. Plants and their associated

microflora evolved in an environment where they were continually assaulted with

a wide array of microbial and plant toxins. Fungal and bacterial toxins are well

known (TeBeest, 1991; Yoder, 1980; Rice, 1974). Certain plants produce “allelopathic” chemicals that suppress the growth of other plants around them

(Putnam, 1985; Durbin, 1981). In addition to plant-produced herbicides, plants

also manufacture a wide range of compounds with adverse pharmacological

effects in herbivores. Rotenone and pyrethroids are plant-produced insecticides.

Coumesterol can alter the mammalian estrus cycles and decrease birth rates.



PHYTOREMEDIATION OF SOILS



79



Taxol can alter mammalian cell cycle (and cure some forms of cancer). The

genetic induction, enzymatic metabolism, and biological effects of some of these

plant components (e.g., flavonoids, isoflavonoids, and coumarins) are well studied (Stafford and Ibrahim, 1992; Cody e t a / . , 1986). Many of the initial steps in

their production from the PAL (phenyl alanine ammonium lyase) have been

cloned and successfully expressed in other plant tissues. This is an active area of

research and other important enzyme classes, pathways, and genes remain, no

doubt, to be discovered.

In many cases, there is a remarkable structural and chemical similarity between a toxic xenobiotic pollutant and either a natural toxin or a specific natural

enzyme substrate. This is not coincidental. Many acutely toxic xenobiotics gain

their “toxic” classification because they interfere with cell processes in ways

similar to natural products. Some agricultural products are specifically modeled

after a natural analog (e.g., bacterial glutamine synthase inhibitors are equivalent

to a commercial herbicide). It is also not uncommon to seek out plant extracts

and test them for bioactive compounds in the discovery of antimicrobial, antiinsect, and phytotoxic products. All of these natural compounds are synthesized

by biological systems, are tolerated by at least some members of the biological

community, and are finally degraded by organisms in the environment. It is

therefore not surprising to plant biochemists, pharmacologists, and traditional

medicine men that plants and their associated root zones have developed significant capacities to metabolize both natural and xenobiotic toxins.

Most of our knowledge about plant-based metabolism of xenobiotics comes

from the development of agricultural pesticides. This technical basis dramatically skews our knowledge base for two reasons. The first is that most studies

have been carried out on pesticides and not on industrial pollutants. Studies on

the uptake, translocation, tolerance or metabolism of industrial pollutants represent only 3% of the literature base (Nellessen and Fletcher, 1993). Second,

pesticides are generally tested on crop plants, most of which have little chance of

survival in truly contaminated soils. Tabulated data again show that 77% of the

studies were conducted on crop plants, and that the metabolic capacity of most

weeds is unstudied (Nellessen and Fletcher, 1993).

Many people unfamiliar with contaminated soils expect that such sites would

be barren of all vegetation. In some cases this is true, however, on most sites

hardy, tolerant, weed species exist. These “volunteers” spread out over time to

establish a general cover at most sites. Often sites that are heavily polluted are

colonized from the edges inward, with the rate of colonization seemingly dependent on contaminant load, physical soil factors, and general cultural conditions.

Many of these sites spontaneously revegetate, the most common exception being

those sites with active weed control programs. This spontaneous revegetation

phenomenon is probably not unfamiliar to those who have spilled oil or gasoline

on the lawn. In cases where spontaneous revegetation does not occur, fertilizer,



80



S. D. CUNNINGHAM E T AL.



loosening up the soil, and water may make dramatic improvements. We have

surveyed many contaminated areas and seen this spontaneous revegetation occurring across many soil types and climates. It is also interesting to note that the

species which seem to come into these areas are nearly unanimously the weed

species coincident with that region, These hardy weed species tend to have windor animal-borne seed distribution techniques and can be found growing in many

of the nutrient-poor, disturbed areas (e.g., road cuts, abandoned fields) throughout the region.

Perhaps not so coincidentally, it is many of these species that are problematic

weeds in farmers’ fields and are therefore specifically targeted in the development of new herbicides. In one author’s experience (S.D.C.) at least half of the

top 10 weeds targeted for new herbicides development can readily be found as

volunteers on contaminated soils in that region. As these casual field observations might indicate, and as the best efforts of hundreds of the world’s pesticide

chemists can attest to, certain weed species are difficult to kill and are relatively

insensitive to chemicals that easily kill crop plants. Given the skewed data base

on metabolic capacity of plants, it appears obvious that phytoremediation would

benefit from additional research in the study of weed metabolic capacity.

The results of agricultural product research also suggest there are wide differences in the ability of plants to metabolize xenobiotics. The backbone of the

multibillion dollar selective herbicide business is based on this differential metabolism. Nearly all modem selective herbicides are selective due to plant metabolism. The tolerant plant selectively metabolizes the herbicide to a nontoxic

compound and remains unaffected, while the nontolerant weed species either

cannot metabolize the compound or metabolizes a nontoxic compound into a

toxic one, thereby committing suicide (Hathway, 1989). There are significant

differences between monocots and dicots in this capacity as well as between

individual genera and species that may be exploited in phytoremediation. Differences in plants’ abilities to metabolize environmental pollutants are also increasingly evident from screening at both the whole plant level (Schnoor er al., 1995;

Hughes and Saunders, 1995) and the cell culture level (Groeger and Fletcher,

1988).

Plant xenobiotic metabolism is remarkably similar to the types of xenobiotic

metabolism that occur in mammalian livers. Relatively lipophilic materials undergo an enzymatic attack, which results in more water-soluble compounds. In

mammalian systems the final disposition is often through excretory routes which

plants lack. In principle, however, both plant and liver metabolic systems can be

divided into the same three phases: transformation, conjugation, and final disposal. The final stage in plant metabolism consists of transport and compartmentalization of the metabolized products into cellular vacuoles, intracellular spaces, or

various cell wall components (Sandermann, 1992). Contributing to the comparison is the fact that two major enzyme systems responsible for liver detoxification

processes are also found in plants: ( 1 ) cytochrome P450 oxygenases, and (2)



PHYTOREMEDIATION OF SOILS



81



glutathione S-transferases. It is on the basis of these comparisons that plants have

been dubbed “green livers” (Sandermann el al., 1977).

Some genes necessary for the degradative capacities in plants are constitutively expressed. Many herbicides are metabolized in some plant species almost

immediately and there is no detectable lag phase. For phytoremediation processes that would rely on these processes functioning, concentration thresholds

would be expected to be fairly broad. In other cases, enzymatic activity is not

constitutive but induced. In these cases a low level of a pollutant may have one

fate in a plant and a higher level might have another. Much like certain microbial

systems, the presence of one toxicant (e.g., toluene) can induce metabolic pathways that will then also degrade other pollutants. In plant metabolism, the

clearest example of a parallel phenomena is in the development of compounds

called “herbicide safeners.” These chemicals are applied either prior to the application of the pesticide or coincident with its application. Their purpose is to

trigger the production or activation of degradative enzymes which degrade the

herbicide before it has a lethal effect on the plant. Most herbicides kill plants by

interfering with a specific metabolic pathway or process. Sublethal quantities of

herbicide or analogs may also trigger this inducible metabolic activities in many

cases. Exogenously applied inducers of metabolic activity are an underutilized

laboratory and field tool in phytoremediation. A review of the area of these

metabolic enhancers and other potential manipulations of plant degradative capacity is provided by Hatzios and Hoagland (1989).

It is widely speculated that phytodecontamination systems might be most

appropriately managed by maximizing the various stress conditions on the plant

(chemical, fertilizer, planting densities, etc.). This is an approach considerably

different than managing a crop for conventional yield purposes. Much more

research in this area is needed prior to making informed field decisions.

Despite all the above discussion of plant degradative capacities, plant metabolic systems pale in comparison to their microbial analogs. Plant systems do not

have as broad a substrate range, nor can they act over as wide a concentration

range, as their microbial counterparts. This can easily be illustrated by comparing the respective abilities of plants and microbes to break aromatic and aliphatic

C-Cl bonds. In plants, there are perhaps four well-documented C-Cl bond

breakage reactions (Hathway, 1989). In microbes there are perhaps a dozen

(Neilson. 1990; Chaudry and Chapalamadugu, 1991). This type of comparison

has prompted the current research to exploit plant-microbial associations and/or

engineer plants for better metabolic activities.



3. Sequestration within the Plant

A xenobiotic compound entering the plant need not necessarily be metabolized

for successful phytoremediation. In some cases, it may be possible to use root

crops with high lipid contents to absorb lipophilic organics from the soil. The



82



S. D. CUNNINGHAM ET AL.



roots would then be harvested and processed. This type of phytoextraction of

certain contaminants from low-organic-matter soils may be possible. One replicated pilot scale experiment used carrots to remediate a DDT-contaminated soil.

Five carrot varieties were grown, harvested, solar-dried, and incinerated. DDT

levels were reported to decrease in a carrot variety dependent manner by 30 to

86% over the controls (McMullin, 1993). The need for harvesting and postharvest processing has economic consequences on the phytodecontamination

scheme; however, it would appear that phytoextraction and root harvesting may

be potentially viable in certain cases. There remain, however, significant questions concerning the general utility of such an approach across soil types and

contaminants.

Beyond phytoextraction, if plant roots could be demonstrated to take up a

pollutant and sequester it into an unavailable fraction, such a process might also

be useful as a basis for phytostabilization. In such a case it would be imperative

that the compound be so tightly sequestered that it would be essentially unavailable even to animals that might feed on the root tissue. This process has been

clearly demonstrated in the case of certain I4C-labeled pesticides that become

irreversibly bound into plant roots. Such residues are resistant to exhaustive

solvent, acid, alkali, and enzymatic extraction protocols. Furthermore, direct

feeding to rats does not result in release of the compound into the animal’s

system and the rat passes the labeled compound through the digestive system

with the other nondigestible fraction of the food (Kahn, 1982). Residues deposited in the lignin fraction of the plant seem to be relatively biologically inert. As

they do not appear in regulatory extraction protocols they are often considered

“degraded.” Additional research is needed to determine the extent to which such

materials are released upon the death and decay of the root.



V. PHYTOREMEDIATION EX PLANTA

A. Ex PLANTAENZYMATIC

EFFECTS

As has been previously stated, plant enzymes can metabolize a wide variety of

xenobiotic pollutants. Plant degradative enzymes are not limited, however, to

functioning only internal to the plant root, stems, and leaves. These enzymes can

also be found in their active forms both in soil and in sediments.

The best characterized of these plant enzymes that occur external to the plant

root are certain oxido-reductases and laccases. More. recently, however, a greater

variety of active plant enzymes has been discovered in sediments far from their

plant source. These enzymes include dehalogenases, nitroreductases, and nitrilases (Schnoor et a l . , 1995).



PHYTOREMEDMTION OF SOILS



83



Some of the best studied oxido-reducatases are the peroxidases. These enzymes have been identified on the external root surfaces of such diverse plants as

cotton (Mueller and Beckman, 1978), wheat (Smith and O’Brien, 1979), cress

(Zaar, 1979), and tomato and water hyacinth (Adler et al., 1994). In tests with

water-borne contaminants the capacity of these plant enzymes to interact with

phenolic, aniline, and certain other aromatic contaminants is well documented

(Adler et al., 1994; Dec and Bollag, 1994). Due to analytical difficulties, similar

reactions are more difficult to clearly demonstrate in soils. Initial research in this

area by a number of labs, however, seems promising, and additional work is

ongoing.

The result of the action of oxido-reductases on pollutants is often the polymerization of the pollutant either onto the root surface or into the soil humic fraction.

Contaminants bound in such a manner are no longer available to most, if not all,

biological processes and ordinary chemical extraction protocols. Of particular

note is the fact that they are not extracted by regulatory mandated extraction

protocols. These polymeric complexes are often referred to as “bound residues.”

The overall enzymatic incorporation into the polymeric humic fraction of soils

has also been referred to as the “humification” process, and is depicted in Fig. 2

as part of phytostabilization. Some researchers in this field would suggest that the

humification process should be listed under the phytodecontamination category,

as the regulatory analytical results suggest. We consider humification a stabilization process, as certain analytical techniques, including some forms of hightemperature thermal extraction and super critical fluid extraction, have been

shown to release some of these bound residues.

There are significant differences between plants at the level of enzyme production. There may also be differences in the release rate and timing (age, season,

stress induced, etc.) although this remains to be tested. Certain plants (e.g.,

horseradish, Armoruciu rusticana) are cultivated for their root enzymatic capacities. Their value as a condiment and in commercial enzyme production is derived

from their peroxidase production. Their potential use in soil remediation is

untested; however, they have shown intriguing possibilities in water decontamination processes (Dec and Bollag, 1994). These investigators are currently

conducting a survey of plant laccase and peroxidase and initial results show

significant variations between plant species.

In addition to humification by enzymes directly derived from the growing plant, fungal symbionts, parasites, and saprophytes, living in conjunction

with the plant and its detritus, produce a wide variety of enzymes which may

be involved in humification (Bollag et a l., 1995). Individually. same of these

fungal species (e.g., white-rot fungi) are specifically targeted to pollutants

(Yadav and Reddy 1993) and are currently being used in some field scale remediations.

Degradation capacities depend not only on the production of these enzymes,