Yr.‘: ; )oux

:u.

. saw

9’

its. II;

V. oak

 

kahuna... .

. II
.. - filo-v
Y. s at)“.

~03...»th 0.1) a: .
In

DE . Pl .r (II

129%.”,3. mass

‘ 2‘

-2».
5.5.x t:
. ., t;

.55 thrill. I
{Ii 1. Evit‘n

"<
vii-[E
1. a..(. {ll}?- .
. 'DID r.>: .vE‘v‘. "
run-.3: ‘ ottll.ol.i§

19'... . 5-C- .
‘3‘?»1. 38...! I

$25.3“.

r
.13.; is.
£-§ (Iv;
I If. r5300} 1

P loll.) -
I; v hiliw.
I ‘

ii...t-
. 'ul‘l~v‘fiv'

I231. av:lv! Yfiiv.
'o. ulf‘l.;~l!vvl.n)‘t '53..“
I .30.... 01.: 221‘- : 00......- I

051.:
If!

1'. al

rtllltv. 1".
(II...)-
’37....

.ikal’tailOII‘al‘t 090‘
(“I 153')er (:07 run!!! “I“

; I .b‘ .O) v '1. Oilfif‘llv‘

{Ht-ISIS

lllllLUlllll

                                                                       

mlllllllllllll

1293 00914 3102

This is to certify that the
dissertation entitled

Responses of Reconstituted Olfactory

Receptors to Diethylsulfide Derivatives

presented by
Menekhem Zv iman

has been accepted towards fulfillment
of the requirements for

 

 

Ph .D . degree in Physiology

Major professor
Wé’ /?? H. Ti Tien, Ph.D.
1 j
r

 

MSU is an Affirmuu'w Action ‘Equu/ Oppurlum’tv Institution 0712771

 

‘—-

 

 

LIBRARY j
Michigan State

University

L _J

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or betore date due.

DATE DUE DATE DUE DATE DUE

l l
l: i-
i A
Jig
_—l——_ '

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity lnflltution
emana-m

 

RESPONSES OF RECONSTITUTED OLFACTORY RECEPTORS TO
DIETHYLSULFIDE DERIVATIVES

By

Menekhe‘m Zviman

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Physiology

1 993

ABSTRACT

RESPONSES OF RECONSTITUTED OLFACTORY RECEPTORS To
DIETHYLSULFIDE DERIVATIVES
By

Menekhem Zviman

The nature of the molecular structures that encode olfactory information has
been a major aim of the study of olfaction. The stereochemical configuration of a
moiety, a part of the odorant molecule, has been proposed to be the elementary
building block of the olfactory information. It was further suggested that an
odorant molecule can contain more than one of these building blocks. Thus, an
odorant molecule can contain more than one substructure each of which is
detected by different receptors.

To investigate this possibility we studied the responses of olfactory
receptors, reconstituted in bilayer lipid membrane (BLM), to three structurally
related odorants. Diethylsulfide, thiophene and diethanolsulfide (thiodiglycol)
were presented to BLMS containing bullfrog cilia membrane fragments.
Diethanolsulfide is a di-alcohol derivative of diethylsulfide. The two molecules
are similar in structure but differ substantially in their odor. Thiophene is a
hetrocyclic molecule that resembles diethylsulfide in its chemical formula as well
as odor. We hypothesized that diethanolsulfide contains the substructure(s)
presented to the receptors by diethylsulfide and thiophene and an additional
substructure not shared with the other two odorants.

To test this hypothesis we presented the odorants to the reconstituted

receptors one odorant after the other at saturating concentrations. The rationale

Menekhem Zviman

for this test is: the reconstituted system will respond to the second odorant only
when the second odorant contains a substructure not shared with the first
odorant. If the first odorant contains all the substructures that the second
odorant presents to the receptors, then the reconstituted system will not respond
to the second odorant. This is because the first odorant will saturate the
receptors.

To establish the saturating concentrations, dose-response relationships
were determined. We found that diethylsulfide had a maximal effect at a
concentration of 7.59 nM, thiophene at 2.16 nM and diethanolsulfide at 195 nM.
Diethylsulfide and thiophene, when presented as the second odorant, did not
elicit an increased conductance of the modified BLM. In some of experiments,
where diethanolsulfide was added after diethylsulfide or thiophene, an increased
conductance was observed. The increased conductance was observed after the
reconstituted receptors did not respond to diethylsulfide or thiophene and in two
experiments after the system did respond to the first odorant.

The inability of diethylsulfide or thiophene to induce a response when
another odorant was already present suggests that all three odorants contain the
same substructure(s) that binds to the olfactory receptors. The response to
diethanolsulfide when it was the second odorant suggests that it contains an
additional substructure which is not shared with the other two odorants.

To test these finding in vivo we compared the ECG responses of intact
epithelia to diethanolsulfide, diethylsulfide and a mixture of them by calculating a
mixture discrimination index. We obtained a value of 1.43. This value indicates
that some of the receptors were activated by both odorants but some were
activated by only one of the odorants. This observation supports the findings

from the reconstituted system.

Dedicated to family and friends whose love and

support enabled me to complete this work

CI.

ACKNOWLEDGMENTS

I thank my major adviser Dr. H. T. Tien and to Drs. R. Bernard, R. A. Fax, W.
L. Frantz and P. J. R. Cobbett, my guidance committee members for guidance
and advise during this study. I also wish to thank Dr. J. E. Chimoskey,
Chairperson and Dr. C. C. Chou, Director of Graduate Studies, Department of
Physiology for encouragement and support. Special thanks are due to Dr. J.
Krier for his interest, advice, and encouragement.

With gratitude I thank my laboratory colleague Dr. A. Tripathy for his help and
friendship; his experience and our discussions made these years bearable.

With a deep sense of debt, I thank my wife Denise, my children Jennifer,
Skotty and Adam for their love, understanding and patience. Lastly, I wish to
thank my parents for their support and sacrifies; their faith in me brought me to

this point; this work is as much theirs as it is mine.

TABLE OF CONTENTS

Page

LIST OF ILLUSTRATIONS ................................................................................. vii
LIST OF TABLES .................................... . ............................................................ xiii
LIST OF ABBREVIATIONS ................................................................................. xvi
INTRODUCTION .................................................................................................. 1
Anatomy ..................................................................................................... 3
General ........................................................................................... 3

Structures and connections ............................................................. 4

Significance for odorant recognition ................................................ 6

The receptor cell ........................................................................................ 8
Cellular Markers .............................................................................. 8

Components of the transduction system unique to olfaction ........... 9

Biochemistry .................................................................................... 9

Electrotonus structure of olfactory receptor cells .......................... 11

Ionic currents ................................................................................. 12

Ion channels .................................................................................. 18

Summary of transduction mechanism ........................................... 21

Implication of anatomical and cellular data .............................................. 24
Proposed theories and mechanisms of olfaction. .................................... 25
Structure-activity relationship ................................................................... 31
Amoore's stereochemical theory ................................................. 32

Specific anosmia ........................................................................... 32

Beet's functional groups ................................................................ 34

Evidence in support of the functional groups model ..................... 36

Mixtures .............................................................................. 36

Chemical modification ........................................................ 37

Summary of structure-activity relationship .................................... 37

Structure of odorants ............................................................................... 38

V

General features of odorants ........................................................ 38

Examples of known odorant SAR ................................................. 39

The odorants used in this study. .............................................................. 41
Hypothesis ............................................................................................... 44
Reconstitution .......................................................................................... 45
Objectives of the study ............................................................................. 47
METHODS .......................................................................................................... 48
General .................................................................................................... 48
Purification of membrane vesicles ........................................................... 48

BLM .......................................................................................................... 52
Chemicals ................................................................................................ 59
Experimental Design ................................................................................ 59
Experiments ........................................................................ 61

Controls .............................................................................. 63

EOG recording ......................................................................................... 68
Analysis .................................................................................................... 70
RESULTS ........................................................................................................... 75
Dose-response curves ............................................................................. 75
Responses to pairs of odorants ............................................................... 83
Frequencies of responses ............ . ............................................................ 99

E06 and MDI measurement .................................................................. 103
DISCUSSION .................................................................................................... 108
CONCLUSIONS ................................................................................................ 126
APPENDIX A Acetone Extraction Procedure ................................................ 128
APPENDIX B Layout of the connection box printed circuit board ................. 129
APPENDIX C l-V measurment program ........................................................ 130
Code Files: .................................................................................. 130

Header files ................................................................................. 167

REFERENCES ................................................................................................. 181

Figure

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

LIST OF ILLUSTRATIONS

Page
Structures of diethylsulfide (A) and l-carvone (8). Both
are viewed from a similar plane (carvone's ring plane) ............ 42
The three odorants used in this study. (A) - Thiophene,
(B) -Diethanolsulfide and (C) - Diethylsulfide. ......................... 43
Comparison between the two isolation methods. Method
A is EtOH/CaZ+. Method B is mechanical agitation. A
and B are wet tissue per animal for both methods. C and
D are amount of protein per animal. A and C are
experiments when method A was used and B and D are

when method B was used. Error bars represent standard

Fusion of vesicles with a planar BLM. A - scheme of a

BLM. B - initial stage. prefusion state (vesicle

attachment). D - osmotic swelling of the vesicle and

fusion. F - final stage: incorporation. ....................................... 56
Current record during reconstitution. The BLM was held

at +40mV. A sharp increase in the current is typical to

fusion of a vesicle(s). Arrows mark vesicle suspension
injection. Channel activity can be seen immediately after

the fusion. ................................................................................ 57

vii

Figure 6.

Figure 7.

Figure 8.

Figure 9.

Figure 10.

Measurement system scheme. I. picoammeter, V.

electrometer and C. capacitance meter. 81 and S2 are
switches. .................................................................................. 58
Chemical structure of the odorants .......................................... 60
Effects of DES and THP on BLM. No significant change

of conductance was onserved .................................................. 65
Effects of THP and DOS on BLM. No significant change

of conductance was onserved .................................................. 66
Effect of DMSO on a BLM. No significant change of

conductance was onserved ...................................................... 67

Figure 11. An illustration of the setup used to record EOG. ........................... 69

Figure 12.

Figure 13.

Figure 14.

Figure 15.

Voltage-current relationships measured in an experiment

to determine a dose-response relationship. The

conductance of the modified BLM was calculated as the

slope of each potential-current relationship. ............................ 71
Voltage-current relationships measured in an experiment

to determine responses to the three odorants. The

conductance of the modified BLM was calculated as the

slope of each potential-current relationship. ............................ 72
Dose-response curve of reconstituted OCM to DES.

Pooled data from three experiments. x2 = 0.509; p =

1.00; E05% = 1.05 nM; ED50% = 2.63 nM; E09596 =

7.59 nM. ................................................................................... 77
Dose-response curve of reconstituted OCM to DOS.

Pooled data from two experiments. 12 = 0.130; p = 1.00;

E0596 = 83.2 nM; ED50% = 126 nM; ED95% = 195 nM. ......... 78

viii

Figure 16.

Figure 17.

Figure 18.

Figure 19.

Figure 20.

Figure 21.

Figure 22.

Dose-response curve of reconstituted OCM to THP. x2 =

0.01; p = 1.00; ED5% = 0.28 nM; ED50% = 0.71 nM;

E09596 = 2.19 nM. ................................................................... 79
Comparison of the Dose-response curves of the three

odorants. .................................................................................. 80
Dose-response of reconstituted OCM to THP. The data

was fit to two subsets of the data. A - Lower few points

were fit using the same parameters as in fig 16 ; B -

Upper subset was fit independently: ED50 = 3.55 nM,

ED5 = 2.88 nM and ED95 = 4.37; C - th sum of A and B. ....... 82
Response to both DES and DOS. The membrane

conductance after fusion was 76.8 pS. .................................... 85
Membrane conductance values from Figure 19 plotted as

the difference between successive treatments (the net

effects of each of the treatments) ............................................. 86
Responses to THP and DOS. The membrane

conductance after fusion was 1.57nS ...................................... 87
Membrane conductance values from Figure 21 plotted as

the difference between successive treatments (the net

effects of each of the treatments). A - THP induced a
conductance increase of 130 p8; B - ATP and GTP did

not have an effect; C - DOS induced additional

conductance change of 80 pS; D - 4-aminopyridin

reduced the conductance by 340 p8; E - 0080;; did not

have an effect. ........................................................................ 88

Figure 23.

Figure 24.

Figure 25.

Figure 26.

Figure 27.

Figure 28.

Response to DOS and no response to a subsequent

addition of DES. The membrane conductance after

fusion was 438 p8 .................................................................... 89
Membrane conductance values from Figure 23 plotted as

the difference between successive treatments (the net

effects of each of the treatments). A - DOS induced 82

ps; 8 - DES changes the conductance by -6 p8; C - 150

nM 4-aminopyridine blocked 39 pS; D - increasing the
concentration of 4-aminopyridine blocked additional 81

pS to a total of 130 p8. ............................................................ 90
Response only to DOS. DES was introduced before DOS

and THP was introduced after DOS, both with no effect.

The membrane conductance after fusion was 126 p8. ............ 91
Membrane conductance values from Figure 25 plotted as

the difference between successive treatments (the net

effects of each of the treatments). A - DES induced a -20

pS change; B - DOS induced a 332 p8 change; C - THP
induced -18 pS change; D - 4—AP induced a -69 pS

change. ................................................................................... 92
Response to DOS without a response to a subsequent

addition of THP. The membrane conductance after

fusion was 147 p8 .................................................................... 93
Conductance values from Figure 27 plotted as the

difference between successive treatments (net effect of

each treatment). A — DES induced a -20 pS change; B -

DOS induced a 319 p8 change; C - THP did not have an

Figure 29.

Figure 30.

Figure 31.

Figure 32.

Figure 33.
Figure 34.

effect; D - 4-amlnopyridine reduced the conductance by

74 pS. ...................................................................................... 94
Net membrane conductances when the only response

was to DOS. THP and DES which were introduced

before DOS had no effect. The membrane conductance

after fusion was 553pS. ........................................................... 95
Membrane conductance values from Figure 29 plotted as

the difference between successive treatments (depicting

the net effect of each of the treatments). A - THP induced

a -16 pS change (note: within the noise level); 8 - ATP

and GTP induced a -31 pS change; C - DES induced a -

2 p8 change (note: within the noise level); D - DOS

induced a 80 pS change; E - 4-AP induced a -55 pS

change; F - CDSO4 induced a -101 pS change. ..................... 96
Potential trace of EOG responses to increasing

concentrations of THP; A - Depolarization; B -
Hyperpolarization. .................................................................. 1 04
Dose-response relationship of EOG to THP. ED50 is

0.513pM; x2 = 0.038; p = 1.00; ED5 = 3.39 nM; ED95 =

6.61 mM. ................................................................................ 105
EOG trace during a MOI of DES and DOS assay. ................. 106
Average MDI of DES and DOS. The three bars depict the
average normalized EOG responses to: DES, DOS and a
mixture of them. Average MDI is 1.43 1: 0.037 (mean :

se.; n = 5). DES and DOS were presented at

concentrations of 1 (M. The responses are normalized

by expressing Ri /{(RDES + R005) I 2}. Average RDES
xi

Figure 35.

Figure 36.
Figure 37.

Figure 38.

is 1.01 1 0.057 and average R003 is 0.99 :I: 0.057

(mean i s.e.; n = 5). .............................................................. 107
Flowchart of the interpretation of the response of

reconstituted OCM to a pair of odorant. The odorants are
introduced one after the other in saturating

concentration. ........................................................................ 1 14
A schematic representation of the three odorants. ................ 118
A schematic representation of the four putative

receptors. ............................................................................... 119
Graphic representation of the analysis of the types of

response in terms of the required receptors. ......................... 123

xii

Table
Table 1

Table 2.
Table 3.

Table 4.
Table 5.
Table 6.
Table 7.
Table 8.

LIST OF TABLES

Page
Ionic currents in Olfactory receptor cells. A - Mudpuppy,
B - Rat, C- Catfish, D - Frog, E - Bullfrog. (-) not present,
(+) present and (nr) not reported. (*) Was observed in
salamander. ............................................................................ 16
Identified ion channels in olfactory receptor neurons. . ........... 19
Evidence relevant to the mechanisms of odorant
transduction. . .......................................................................... 23
Proposed theories for olfaction and odorant structure. .......... 27
Expected responses to different pairs of odorants ................... 61
Odorant introduction protocols ................................................. 63
Effective dose (ED) values for the three odorants. .................. 81

Frequencies of all the responses to odorants. The
responses are not divided in relation to other odorants
that were tested. The responses are further subdivided
into those that were produced when the odorant was
tested first. Responses to DOS are further subdivided
into those that were produced when DOS was presented
as the second odorant. DES and THP did not elicit a
response when presented second. The last two rows

total responses for the three odorants: a subtotal of the

xiii

Table 9.

Table 10.

Table 1 1.

responses to the odorants when presented as the first

odorant and the grand total of all the responses. (N

tested) are the total number of experiments where the

odorant was tested; (?) are the number of unclear

responses to the odorant; (N clear y/n) are the number of
experiments where a clear response or no response

were observed. (Yes) are the numbers of responses to

the odorant. (No) are the number of no response to the

odorant; (Frequency) the proportion of positive

responses from these experiments where clear

responses were observed. .................................................... 100
Responses to pairs of odorants. ( - -) No response to

either. (-+) No response to the first but response to the

second odorant. (+-) response to the first odorant without
additional response to the second odorant. (++)

Response to the first odorant and additional response to

the second one. (B) stands for 'followed by’ .......................... 101
Approximate confidence intervals for the proportions

from Table 8. Confidence level is 90%(two tail test, a of

each side is 5%). DES, THP and DOS indicate the total

number of tests. DES1, THP1 and D081 indicate the
experiments where these odorants were presented first.

DOSZ indicates the experiments where DOS was the

second odorant to be presented. ........................................... 112
Proportions of independent responses to the three

odorants. ................................................................................ 1 18

xiv

Table 12.

Frequencies of various responses from table 9 arranged
according to the classification in fig 38. Total number of
experiments which tested DES is 27 for A to C and 19 for
D to F; Total number of experiments which tested THP is
18 for A to C and 20 for D to F; Data in the columns are
the number of observations that fit the type of response
(A to F) and next to it the proportion of that response
from the experiments it was tested. The pooled
proportions (i.e., probabilities) of the proportions of THP
and DES are presented in the right column (pooled by
combining the data from the experiments where DES

was used with those where THP was used) ........................... 121

4-AP
AC
AE
AON
BLM
DES
DOS
ED
EOG
IBMP
MDI
OB
OBP
OCM
OE
OMP
ORC
OT
PLC
SAR

SBTPE

LIST or ABBREVIATIONS

4-Aminopyridine.

Adenylate cyclase.

Acetone extract of SBTPE.
Anterior olfactory nucleus.
Bilayer lipid membrane.
Diethylsulfide.
diethanolsulfide.

Effective dose.
Electroolfactogram.
3-isobutyl-2-methoxypyrazihe.
Mixture discrimination index.
Olfactory bulb.

Olfactory binding protein.
Olfactory cilia membrane.
Olfactory epithelium.
Olfactory marker protein.
Olfactory receptor cell.
Olfactory tubercle.
Phospholipase C .

Structure activity relationship

Soy bean total phophotide extract

xvi

THP Thiophene.

xvii

INTRODUCTION

Olfaction is one of the chemoreception systems. In land animals, it is the
system that detects and identifies the chemical environment that surrounds an
animal. Sensory systems detect and transduce particular types of external
energy into neural code. Some features of this external energy are a meaningful
stimulus to the sensory system. The features of the stimulating energy in
olfaction are not clear. Stimulus in olfaction can be viewed as the result of
interactions between a compound and its receptor(s) or as that substructure(s)
of the odorant that is detected by the olfactory receptor cell(s). The sensory
signal is assumed to be the result of a ”formation of a reversible non-covalent complex”
between a proteineous receptor and an odorant molecule (90).

The transduction of a stimulus to neural code is first manifested as a change
in the conductance of the receptor cell membrane. This change can be used as
a probe for the interaction between the receptor and its ligand -- the odorant.
While other techniques can be employed to study the odorant-receptor binding,
using the conductance change ensures a focus on these receptors that are
significant in the context of the normal function of the receptor cell. The coupling
mechanisms of receptors to the conductance of the membrane vary among
receptor types. The choice of technique to monitor the conductance change
dictates which types of coupling mechanisms will be relevant. This study takes
advantage of the unique feature of reconstituted plasma membrane fragments:

isolation of the processes occurring at the membrane from any cytoplasmic

1

2

influence. This feature facilitates the study of specificity of olfactory receptors
without the complications arising from regulatory processes which depend on
cytoplasmic mechanisms.

One of the paramount questions about the olfactory system is the physico-
chemical nature of its stimuli. Many mechanisms were proposed for the
transduction of odorants'. In this study receptor dependent mechanism is
assumed. Thus, the questions asked in this study relate to the odorant-receptor
relationship rather than the odorant-transduction mechanisms. Furthermore, the
choice of coupling mechanisms that are restricted to the boundaries of the
plasma membrane narrows the transduction mechanisms to those that involve
direct coupling.

The knowledge of the physics of the stimulus enables the researcher to
understand the relationship between the stimulus and the response of the
receptor cell. Very little is known about the nature of the olfactory stimulus
molecules. Actually we know a lot more about the transduction machinery than
about the stimulus itself. Theory suggests that odorants contain one or more
substructures that are detected by the receptors in the olfactory epithelia (10,
11, 33, 89, 90, 91, 96). Each of these substructures of the odorants is a unique
informational element (an osmophore). The aim of this study was to provide
evidence in support of the theoretical prediction that a single molecule can be
more than one unique olfactory informational element (the goal is described in
detail in page 47).

The background for this study involves mainly the current understanding of

structure-activity relationship of Iodorants and the transduction mechanisms.

 

' Three terms are used frequently in the context of olfaction: odorant, odor and smell. In the
study of olfaction these terms are used as follows: Odorant is a chemical compound, odor is the
coding of the odorant (including its perception) and smell is the perceived sensation of one or
more odoriferous substances.

3

Olfaction has been studied using a variety of techniques. Some other aspects of
the olfactory system are reviewed. These issues, like anatomy and the receptor
cell biology provide the background needed for the consideration of sometimes

conflicting literature.

Anatomy

I will first describe some of the anatomy and its relevance to odorant
recognition.
Game!

The chemosensory system is traditionally divided to taste and olfaction.
Olfaction comprises three systems: (a) a nasal chemosensory system, (b) the
vomeronasal system and (c) the other chemosensory systems (53, 116, 143).
Most of the knowledge about olfaction comes from studies of the nasal
chemosensory system. The other chemosensory systems were termed "the
common chemical sense". The common chemical senses are divided into
trigeminal nerve and cutaneous chemoreception. Some interaction is reported
between the common chemical sense and both olfaction and taste. The stimuli
sensed by these systems are usually irritants (116). Compared to the common
chemical sense system the vomeronasal system (vomeronasal organ,
Jacobson's organ) is closer related to olfaction. Its structure parallels that of the
olfactory organ. The vomeronasal system is thought to play a major role in the
perception of chemical stimuli of a behavioral relevance including feeding, social
behavior and reproduction (143, 115, 53). This study utilizes epithelia from the

olfactory organ only.

Structures and connections
A simplified scheme of olfactory information transmission can be depicted as

follows (53):

Sensory Olfactory Primary Further

Odorant :> epithelium. => bulb. => olfactory :> cortical and
(receptor (mitral cortex. sub-cortical
cell) cells) (pyramidal areas

cells)

The olfactory system can be divided into several hierarchical levels. Initial
events occur at the olfactory epithelia (OE). Olfactory epithelia contain among
others receptor neurons (ORC) who's axons form the olfactory nerves. ORCS
terminate in the olfactory bulbs (OB). The OBS contain local circuitry and project
to few areas of the brain. The hyppocampus, amygdaloid complex, tegmentum
and the striatum receive projections from the bulb. The bulbs also project to the
olfactory turbencles (OT) and the anterior olfactory nuclei (AON). The AON in
turn projects to the contralateral bulb. The OT projects to the thalamus and from
there to the neocortex. The amygdala projects to the hypothalamus and the
tegmentum (22). The olfactory bulbs are thought to project directly to the
hypothalamus. Thus the olfactory system projects to cortex, to visceral centers
and directly to the limbic system.

Olfactory mucosa lines a large part of the nasal cavity. The olfactory mucosa
contains the OE and a layer of mucus. The epithelium (the cellular part) consists
of two distinctively separate layers: the "neuroepithelium proper" and the lamina
propria. A basement membrane segregates the two layers. The neuroepithelium
proper is a columnar epithelium that contains three cell types: ORCS,
sustentacular cells and basal cells. Tight junctions between ORC dendrites and
the supporting cells are impermeable to charged molecules (66). The lamina

propria consists of highly vascular and glandular connective tissue. The lamina

5

propria contains Bowman's glands, nerve fascicles (axons of the ORCS), blood
vessels and submucosal glands (5, 53, 66, 47).

The mucus covering the OE is secreted by the Bowman's gland and the
sustenacular cells (in some animals) (47). In amphibians, Bowman's glands
secrete a water serous secretion and the sustenacular cells secrete a viscous
mucus. The mucus layer is normally about 35pm thick and contains two layers: a
thicker layer and a watery layer. The distal segments of the cilia lie at the
interface between these layers, about 5pm below the mucus surface. Mucus is
composed of a mesh-like network of glycoprotein fibers. This network forms a
gel lattice in which water containing the dissolved components of the mucus fills
the cavities (49). The mucus sweeps over the surface of the OE at a rate of 10 to
60 mmlmin (48). Cation concentrations as measured in frog olfactory mucus are:
Na+ 52.7, K+ 10.6 and Ca2+ 5.4 (in mM). These concentrations are similar to
those derived from secretory tissue compared to those found in the extracellular
fluid (59).

The first relay station in the olfactory system is the olfactory bulb. The most
noticeable structure in the OB is the glomerulus. It is a Spherical structure
containing dendrites and terminals and surrounded by glia. Each OB contains
the terminals of ipsilateral ORCS. The 08 also contains intemeurons and output
neurons. The projections of the OE to the OB are largely topographically
organized. The 08 has two output cells: mitral and tufted. The dendrites of both
output cell types are located in the glomeruli where they receive synapses from
the terminals of the ORCS. Tufted cells do not leave the 08 but rather project to
distant areas in the bulb. Data from the rabbit suggest that tens of millions of
ORCS terminate inside around 2,000 glomeruli. This should be compared to

about 70,000 mitral cells and about 16,000 tufted cells. The input to output ratio

6

indicates large convergence. 08 contains few types of local circuits ranging
from intraglomerular to extrabulbular (1 10, 53).

Primary olfactory cortex can be defined as the cortical region that receives a
direct axonal input from the main olfactory bulb. These cortical areas are: the
anterior olfactory nucleus, the piriform cortex, the olfactory tubercle, the cortical
part of the amygdala and the lateral entorhinal cortex. The accessory olfactory
system that receives its input via the vomeronasal organ has distinctively
different cortical areas. The different cortical areas are interconnected by an
extensive system of associational connections. Axons from the olfactory bulbs
terminate at the superficial layers of the cortical areas. The superficial layer
contains the dendrites of pyramidal cells. Projections from the olfactory bulbs
lack topographical organization although a complex pattern was suggested.
Most likely, the lateral olfactory track is heterogeneous as far as the
neurotransmitters used. The olfactory cortex projects to orbital neocortex, the
mediodorsal and submedial nuclei of the thalamus, the lateral hypothalamus,
amygdala and hyppocampus (98, 53).

Significance for odorant recognition

Some non-receptor factors influence the recognition of an odorant. The

 

factors that need to be considered span from the physics of the air flow in the
nasal cavity to the neuroanatomy of the higher olfactory centers. Here are a few
examples concerning their usage in the study of the olfactory system.

One of the factors that is commonly used in the study of olfaction is the
sensitivity of the system to an odorant. The sensitivity is studied by asking
testers to evaluate the intensity of odorants and deducing threshold
concentrations. Apfelbach et al. (7) showed that the sensitivity correlates the
convergence ratio of the primary to secondary neurons. Difference in

convergence ratio between ORC'S with different specificity must be taken into

7

account in any comparison. Since the Specificity of ORCS is not known the
comparison between perceived thresholds of different odorants must be used
cautiously.

In air breathing animals, odorants travel from the inhaled air to the surface of
the ORC cilia. This path transverses the mucus layers that cover the epithelia.
While most odorants are hydrophobic, the mucus is an aqueous solution.
Olfactory binding protein (OBP) which is present in high levels in the mucus
interacts with the odorants. At least two roles have been proposed for OBP:
removal of odorants and facilitation of the diffusion of odorants (94, 93).
Because of the interaction of odorants with OBP it is difficult to relate threshold
concentrations from experiments that used odorants in vapor phase and intact
mucosa to those performed in vitro where the mucus has been removed.

Since the odorants have to cross the mucus layer, the thickness of the layer is
important. As described by Getchell and co-workers (47, 49) the thickness of the
mucus layer, the rate of its flow and its viscosity are not constant. Actually, it is
influenced by the olfactory information. The mucus has a role in separating
mixtures of odorants and delivering different odorants to different areas of the
OE. The OE is believed to have some spatial organization of specificity (76).
Therefore, the odorant receptor interaction is influenced to some degree by the
status of the mucus layer.

There are many more factors to consider in this complicated path of the
stimulating molecule and the perceived odor. When analyzing the relationship
between the results of different experimental preparations, it is important to bear

in mind the complications arising from these factors.

The receptor cell

Olfactory receptor cells continuously differentiate from the basal cells of the
DE. The receptor cells are bipolar neurons with dendrites reaching the surface
of the OE, soma near the basement membrane and terminals in the OB. The
axons of the ORCS form the olfactory nerves. Olfactory receptor cells have a
non-branching dendrite that ends with an enlargement (olfactory knob) bearing 5
- 20 cilia. These long cilia (30-200um) lack dinein and thus are immobile (in most
animals) (5, 66, 121, 53, 47). The olfactory knob may have microvilli as well
(53). The receptors for the odorants were shown electrophysiologicaly (42, 44,
73) and biochemicaly (92, 114, 119, 12, 65, 103) to be in the apical and the ciliar
membrane.
Cellular Markers

Some unique markers were found for ORCS. Olfactory marker protein (OMP)
is considered to be a selective marker for a mature and functioning ORC. OMP
is a cytosolic protein of around 18.5 - 18.7 kDa (162 amino acids) (104, 53). The
animo acid sequence of OMP does not provide a clue for its function (121) but it
is restricted to the olfactory neurons and their terminals (9). OMP appears a few
days before the appearance of tyrosine hydroxylase in the target dopaminergic
neuron in the bulb (79). Carnosine is another ORC marker. It is a dipeptide ( B-
alanyl-L-histidine) which is found in relatively large quantities in ORCS. Both the
camosine synthase and carnosinase are found in the ORC. The presence of
these enzymes was shown to be dependent on the connection to the OB.
Carnosine was suggested to be a neurotransmitter. Lack of evidence to support
the role of neurotransmitter together with evidence of the presence of

carnosinase in glandular cell, seems to reject this hypothesis (53, 5).

9
Qommnents of the transduction system unigue to olfaction

Few components of the transduction mechanism were found to be proteins
unique to ORCS. An olfactory GTP-binding protein (Golf) was identified and
sequenced (57). It was shown to be more abundant than Gs in ORCS as well as
restricted to these cells (58). A glycoprotein termed gp95 was found to be the
most prominent component of the cilia membrane proteins (about 25%). The
function of gp95 is not known. Because gp95 has both a constant and a variable
domain, it was suggested as a candidate for the odorant receptor (24, 67, 68). A
specialized adenylyl cyclase was found in the cilia of ORCS of a rat (8).

Olfactory receptor cells are the transducers of the olfactory stimulus to neural
activity. As a transducer an ORC has to have some mechanisms to convert and
encode the odorant signal. ORCS contain odorant receptors that are linked to
ion channels. Several membrane proteins were suggested as putative receptors.
The following is a list of several of these proteins. Two proteins of 61 kDa were
identified in the dog using monoclonal antibodies. A dimer comprised of 88 kDa
and 55 kDa proteins was identified in the rat using polyclonal antibodies. The
dimer is a glycoprotein (similar size glycoprotein was found in rat, mouse,
guinea-pig and hamster) (140). A 95 kDa glycoprotein was found in the frog
(homologous proteins from: toad, rat and cow) (67). Membrane proteins encoded
by a family of genes were also suggested (rat)(17). Three types of linkage
between the receptor and ion channel have been observed: cyclic nucleotides,
IP3 and odorant-activated channels. The details of both the channels and their
respective activators will be discussed below.

Biochemistry

Biochemistry of the olfactory receptor cell has advanced considerably in a

short time. Some of the key components. were isolated and cloned. In a

preparation of frog olfactory cilia, Lancet and co-workers found odorant-

1O

activated adenylate cyclase (AC) (92). A mixture of four odorants: citral, I-
carvone, 1,8 cineol and n-amyl acetate (2.5uM each) induced an elevated cAMP
level. The activation required the presence of GTP. Stimulation of the enzyme by
GTPyS eliminated the effect of the odorants. This was the first evidence for the
involvement of second messenger systems in olfactory transduction. Since then,
a large number of researchers confirmed and expanded the understanding of
this pathway. Olfactory adenylate cyclase was shown to be active in: frog (92),
bullfrog (119), rat (8, 114, 14, 13), salamander (40) and cows (71 ). In the rat,
Ca2+ at physiological concentration inhibits the activity of the AC (114). AC was
isolated and incorporated into liposome membrane (4). A recently cloned bovine
AC was shown to be localized at the cilia region of rat olfactory epithelia. This
AC is a new type that has a low basal activity (8, 71).

For an AC to be part of a transduction system it has to be coupled to a
receptor and cAMP has to cause changes in the cell conductance. Both the
additional parts were identified. Attempts to isolate the olfactory receptor protein
date to mid 1960's. To date quite a variety of proteins were isolated. Recently a
multigene family that codes transmembrane proteins was identified to be unique
to olfactory cells. Features of the encoded proteins support the possibility that
these are olfactory receptors. These proteins bear structural resemblance to a
number of G protein-coupled receptors ('17, 18). The receptors are coupled to
the AC by a GTP binding protein. An olfactory neuron specific G protein was
identified. It is an analog of the stimulatory Ga. The Golf was shown to be
present only in olfactory neurons (57). Within the olfactory neurons it was shown
to be essentially the only Gs in the cilia (58).

An activated adenylate cyclase elevates the concentration of CAMP in the cell.
This in turn must be translated to a change in membrane potential. A cyclic

nucleotide ion channel was discovered by Nakamura and Gold (85). The

11

channel resembles the retinal outer rod cyclic nucleotide-gated channel both in
activity and structure. Several laboratories have cloned and expressed the
channel (rat (28), catfish (51) and bovine (61)). The channel is cation-selective
and is blocked by Ca2+ at negative membrane potentials.

Another second messenger system is the IP3 system. Boyle et al. (12) found
phospholipase C (PLC) activity in isolated catfish olfactory cilia in response to
odorants. The response required GTP, and 90% of the hydrolysis product was
IP3. The same was found in isolated olfactory cilia from rat (14, 13). lP3 is
known to induce an increase of cytoplasmic Ca2+ from intracellular source or by
activating a plasma membrane Ca2+ channel. An IP3 gated Ca2+ channel was
identified in olfactory cilia of catfish (101).

Electrotonus structure of olfactory receptor cells

Once the transduced signal reaches the ion channels a generator potential is
created. The relationship between the generator potential and the neuron's
output depends on the electrotonic structure of the cell. The electrotonic features
of ORC are different from most neurons. The cilia of the ORC are bathed in a
mucus that has higher concentration of K+ compared to the extracellular fluid
below the tight junctions (59). Since the cilia membrane constitutes about 25% of
the ORC membrane, the potential of the cilia strongly affects the soma and
dendrite membrane potential. The cilia have high specific membrane resistance
compared with the soma-dendrite membrane. The high resistance possibly
arises from presence of only a few K+ leak channels (111, 112, 46), resulting in
the K+ conductance almost exclusively in the soma (73). The soma's excitability
is low, since at the resting potential it is mostly inactivated (46). The higher
resistance of the cilia (the current source) together with the low excitability of the

soma result in efficient injection of current to the axon hillock (46). A proposed

12

model for the electrotonic structure of the ORC supports the observation that
ORCS respond to even one odor induced channel opening (74).
Ionic currents

Ionic currents in ORCS were studied using various preparations. The types of
preparations range from whole mucosa to a culture of rat ORCS. Here is a brief
account of some of the currents. Table 1 is a summary of the characteristics of
vertebrate ORC ionic currents. '

In mudpuppy isolated ORC held in whole cell voltage clamp configuration,
the following currents were observed: Na+ current, Ca2+ activated K+ current
that was blocked by Cs“, delayed rectifier K+ current that was blocked by TEA,
Ca2+ sensitive cation current and Ca2+ insensitive cation current . The
following currents were absent: Cl' current, inward rectifying K1” current and
transient K1” current (29). In a similar preparation both depolarization and
hyperpolarization were observed in response to odorants (30).

In rat isolated ORCS, cell depolarization induced by addition of K+ to the
external medium caused a rapid increase in [Ca2+ 15 throughout the cell while
odorant induced increase in [Ca2+ Ii was generally localized at the apical end
(100). The following currents were observed in rat ORCS in culture. A TTX
(100nM) sensitive Na+ current was found. This current activated at about -60
mV, peaked at about 15mV and reversed at about +35mV. It inactivated at -30
mV with half inactivation at -63mV. A high threshold Ca2+ current was recorded.
This current was Similar to the type L current. It activated at -35mV, peaked at
+10mV, reversed at +60mV and blocked by 100pM Cd2+ or 1uM nifedipine. A
delayed rectifier K+ current was recorded. The current activated at -30mVand
was blocked by 25mM TEA. It was not blocked by 4-AP or Cd2+ and did not

inactivate. No transient K+ current was observed (129).

1.3

The following currents were observed in isolated catfish ORCS. A TTX (1 uM)
sensitive Na+ current. This current activated at -70mV to -50mV and had
reached a maximum at -10 mV. The current inactivated. Half maximal
inactivation was recorded at -62mV and complete inactivation at -30mV. A Ca2+
current that is blocked by 2mM 002+ or 5uM nimodipine was observed. This
current was similar to the L-type current. It activated at -40 to -30mV and
reached a maximum at OmV. A Ca2+ activated K+ current was found. It was
reduced by 2mM C02+, reached a maximum at +30mV and was blocked by
20nM CTX or 250nM apamine. A transient K+ current was found. It was similar
to A-type current. It inactivated at -40mV, and was blocked by 10mM 4-AP. A
sustained K+ current, blocked by 10mM Ba2+ or 20mM TEA or 10mM 4-AP, was
found. This current resembles the delayed rectifier current. A K+ inward rectifier
current was found. It was activated at -80mV, was not suppressed by replacing
K"’ by Cs". It was blocked by 10mM Ba2+ in both sides. An odorant induced
Ca2 + dependent current was recorded. The current was not dependent on ATP,
it reached a maximum at -60mV and was decreased at negative membrane
potentials. An odorant induced outward rectifier current was recorded. The
current did not respond to amiloride, it was blocked by extracellular Ba2+ and
the response to odorants was abolished by addition of forskolin. An odorant
induced current was observed. It had a linear conductance. The current had a
reversal potential at -15mV and was blocked by 500p.M amiloride (83).

The following was observed in frog or salamander isolated ORCS. Na+
currents: in frog the current was blocked by ‘I'I'X (1 uM) but in salamander a TTX
insensitive current (blocked by Co”) was observed. A K+ delayed current was
found. It was blocked by Cs"; or TEA. A transient K+ current was recorded. It
was blocked by 4mM 4-AP, was almost inactivated at membrane potentials

above -40mV and was similar to the A—type current. A Ca2+ (Ba2+ ) current was

14

recorded. It was a voltage dependent current. An odorant-induced inward current
was observed. It was followed by long hyperpolarization. A cAMP induced
inward current was observed. It was not blocked by TTX and decayed to
baseline. Another odorant-induced current was recorded. It had a reversal
potential around +10mV (128, 127). In frog isolated ORCS and in whole mucosa
under current clamp , BrcAMP and odorants induced two-phase response (first
lower potential and than positive potential). This current was blocked by
Amiloride. Phorbolester ester causes single phase positive potential response
(108).

Electroolfactogram (EOG) was used to identify ionic dependencies of the
ORC response to odorants. In the frog, EOGs were measured in the presence
of different ionic environments. It was found that Inorganic cations inhibit the
trans-epithelial potential in an order that correlates with blockage of a calcium
channel (La3+ > Zn2+ > Cd?+ > Al3+ > Sr2+ > Co2+ > Ba2+ > M921"). Diltiazem
and verapamil produced a dose dependent trans-epithelial potential inhibition.
Dihydropyridines (nicardipine and nifedipine) did not block the trans-epithelial
potential. Calmodulin antagonists inhibit the trans-epithelial potential but not in
the order of their potency (trifluoperazine (TFP), chlorpromazine (CPZ) and N-
(6-aminohexyl)-5-chloro-1-napthalenesulfonamide (W-7)). This evidence
suggests that the trans-epithelial potential depends on a conductance through a
calcium channel and that calmodulin is not involved directly in the generation of
the potential(142). Extracellular calcium was found to be essential for the trans-
epithelial potential. CD2+ and La3+ antagonized Ca2+ competitively. Cd2+
depressed the trans-epithelial potential irreversibly. The responses that
depended on the presence of Ca2+ and Ba2+ were found to be additive. 002+
blocked only the Na“ and Ca2+ dependent response. These observations

suggest the presence of a non selective-cation channel (70).

15

The following currents were identified in bullfrog ORCS using a whole cell
voltage clamp configuration. A voltage dependent Na‘“ current that activated at
-40 mV, reached a maximum at —20mV and had a reversal potential of +40mV.
TTX suppressed only 80-90% of the Na+ current. A Na+ TTX insensitive current
was part of the transient inward current. A Ca2+ current that activated at -20mV,
reached a maximum at +10mV and reversed at +40mV. This current was
suppressed by substituting Ca2+ with 002*. A delayed rectifier K+ current that
was blocked by external 20mM TEA or when internal K+ was replaced by Cs+. A
Ca2+ activated K+ current that was blocked by 20mM TEA or when K+ was
replaced by C51". A cyclic nucleotide induced current. The effectiveness of the
nucleotides was: cGMP > cAMP >> cCMP > clMP >> cUMP >>> cTMP. The Ion
selectivity of current was: Na’r > K+ > Li+ > Rb” >> Cs"’ > choline > TEA (1.0 :
0.83 : 0.68 : 0.48 : 0.47 : 0.16 : 0.09). It was measured in the absence of Ca2+.

Ca2+ could also contribute to the current (122, 123).

16

Table 1 Ionic currents in Olfactory receptor cells. A - Mudpuppy, B - Rat, C-
Catfish, D - Frog, E - Bullfrog. (-) not present, (+) present and (nr) not reported.
(‘) Was observed in salamander

 

 

 

 

 

 

 

 

 

 

 

A B c D E Features Blockers
KT'
Transient - - + + nr lnactiveat-40.Similerto 10rnM4-AP.
'A'current.
Delayedrectifler + + nr nr + Activatesat-SOmVNot A-TEA
bloskedbyll-APDT 8-25"“;2?
od”. C-IOrnM *,2omMTEAor10mM4-AP.
E-20rnM TEA (ext). 06 lnsteedofK’Gnt).
lnwerdrectifier - nr + nr nr Activateset-80mV.Not c-IomMBazt.
suppressedbst".
Outwardrectltier nr nr nr +' nr Ca2+,TEA
Ca2*ectlveted + nr + nr + Max. +30mV. A-Cs‘.
E-20mMTEA(ext.Ce*InsteedofK’(lnt).
C-Reducedby *.2onMcrx,2sonM
apamine.
Na":
TTXsensitlve + + + +‘ + B-Activeteset-SOmV, B-IOOnMTTX.
max.et-15mV,reversal C-IpMTTX.
lt+35mV. D-mMTTx.
C-activateeet-70to- D- IpMTTX.
50rnV,max.-10mV.
lnectivetion:helfmax.at—
30rnv,completeet-
62mV
E-Activateset-40rnv,
max.et-20mVand
reversesat+40mV.
TTXIneeneltive nr nr nr " + D'-Co2+

 

 

 

 

 

 

 

 

 

Table 1 (cont'd)

17

 

 

 

 

 

 

 

 

 

 

 

A B c D E Features Blockers
Caz":
D nr + + + + B-Activateeat-SSmV, B-100pM +‘or1uMnifedipine.
max.et+10mVreverees c-2 orSpMnimodiplne.
WW et+60rnV.SlmllertoL- E-C *Instadorcét.
mm” twe-
C-Activateeat-40to-
30mV,max.et0mV.
Similar to L-type.
E-Activatesat-ZOmV,
max.et+10mVreversee
et+40mV. ‘
Others:
Cyclic nr nr nr +' + D-NotblockedbyTl’X.
nucleotides E-Noneelectivecetion.
Odorant
induced:
Inward + nr + +' nr C-lineer conductance C-150uMamiloride.
reverses at -15mV. D ~ amiloride.
D-Salamanderreversal
pctential+10mv.
CeZTdepermnt nr nr + nr or C-thdependenton
ATP.Max.et-60mV
reducesetmorenegative
potentials.
Outwardrectifier nr nr + nr nr C-Doeenotreepondto C-Be2*(ext).
amilorideForskolin
abolisheethereeponse
toodorant.
Outward + nr nr nr nr

 

 

 

 

 

 

 

 

 

18

The data on ionic currents leads to a few points:

. Using the present techniques, it is difficult to separate the ciliary currents
from somatic currents. In other words, the separation of the generator current
from the response current can not be done easily.

. ORCS have a variety of currents. It is difficult to isolate a particular current in
response to a stimulus.

. Both hyperpolarization and depolarization have been observed as a response
to odorant stimulation.

. 002+ has few effects: it blocks some of the EOG response (non selective
cation), it blocks a TTX insensitive Na'I' current, it blocks a Ca2+ current and it
reduces a Ca2+ activated K+ current.

. 4-AP blocks a variety of K+ current types.

Ion channels I
A variety of ion channels was identified in ORCS. Although most of the data

comes from the soma some channels were identified in the dendritic knob or by

reconstitution of ciliary membrane.

Odorants activated ionic channels directly. In the rat, nanomolar
concentrations of diethylsulfide as well as carvone activated a K+ selective
channel (139, 134). In the bullfrog IBMP and citralva activated a cation selective
channel (63). In both cases ATP and GTP which are required for adenylate
cyclase activation, were not present. In ORC phospholipase C is thought to be
activated via a G protein and thus requires GTP.

Along with the presence of odorant activated channels two observations
should be pointed out: '

. The dendritic knob membrane contains both large conductance Cl‘ and Ca2+
channels.

. The cAMP activated channel Shows non-linearity in the presence of Ca2+-

19

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 2. Identified ion channels in olfactory receptor neurons.
Channel Conductance Selectivity and Activated by Blocked by Aninel and
features References
Odorant 62pS K+ 25m 0.125uM 4»AP Rat.
sensitive. 20mM NaCl, 20mM Diethylsulfide or (139,134)
KCI, 2mM CaCIz. 50nM (-)-Carvone
Odorant mitive Multiples of 35pS. Na+ a K” 40nM IBMP Not Identified Bullfrog.
0.2M NaAcetate. IOOnM Cltralva (63)
Kt eelectlve 1eops K” » Rbt » ctr?” Not identified "
200mM KAcetate NH‘ » Nat ,
Ca
Monovdent 40ps Na+ a It+ Not Identified Not identified '
calorie
Inward notifier 27pSi6pS K+ V-dep Salamander.
(some) (128)
Linear 34ps. 120mM K*. K*. Reversal ,,
conductance potenfial -52mV. (126)
Not armed 40pS and BOpS. Not identified Not identified Not identified Frog. (62)
com NaCl, SDmM
KCl, 5th HEPES.
mu rectifier 27pS K‘ Ce or TEA "
Net termed 19p$ measured. 29pS Not identified V-dependent. Rat. (74)
(mud rectifier Ie calculated.
suggested) 145mM KCI, 2mM
CaCIz, etc.
Not termed 35ps measured, 52pS Not Identified -
calcuated. 145mM
KCI, 2mM CaCI2, etc.
Transient I<+ 17 and 26pS K" > Nat V-dep, and (Ktjo Rat (75)
Ca” activated 190pS Matteo.
K" Maue and
Dionne
1987.
CI‘ 13098 "
ea2+ activated 133ps. 145 mM K”. 0152*. and Vm Intracellular Mouse. (80)
K". Some and SOmM Ce’.
mm. 10% ' 0.1_ Nd 4-AP
Not TEA

1.0m ti.

 

 

 

 

 

 

 

 

20

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 2 (cont'd)
Channel Conductance Selectivity and Activated by Blocked by Animal and
features References
Ca2” activated 82pS. 145 mM K”. Less sensitive than Intracellular '-
K". Sorna and the was 50rnM Cat.
dendritic knob. channel.
l<+ 21 pS v sensitive. Not Cs". ,
K’. Cell attached. 45pS Rapid Inactivates '-
depolarwation
L-type 29.415.3pS . 55mM 13a”2 » K“ a 5-10uM Catfish. (99)
BaCl2 Nat Nlmodiplne
V-dep Ca2+. 16pS. lsotonic BaClz. Depolarization Inactivates '
Cl‘. Some and 211pS. "
dendritic knob.
cAMP gated. 3555pS average Na" ; K+ CAMP Divalent cation Catfish. (16)
43.6pS . 85mMNaCI, inthetransslde.
12.5 mM KCI, 1.6mM The blockage is
MgCI2, 0.25mM voltage
CaClz, or 100rnM dependent
KCI.
c-AMP gated 15pS. 140mM NaCl. Not identified CAMP ( = 3.86 Caz” at Frog. (43)
, 11M). «:6 ( K". = narrative
1.42m), cpt- potentials. Only
cAMP ( Km = 0.17 when outside.
(till). 0600, amiloride
and dlltlazem.
CAMP gated 70pS. Not identified cAMP. Might be Porcine heart Rat. (136)
30mM NaCl, 30th phosphorilated in protein klnase
KCI, 2mM CaClz the presence of Inhibitor.
ATP to increase
mean open time.
CAMP gated 32pS Not Identified CAMP ( > 5 uM) 50mM BaCIz Rat. (133,
55mMNaCI, 4 mM 133)
KCI, 2mM MgClz.
1mM CaCIz.
CAMP gated Not determined Not identified Not identified Ca2+ blocks at Toad. (85)
monovalent the negative
cation potentials
IP3 gated 79pS, 55rnM Ba“2 Ba‘2 » K’ » sum IP3 10pM Catfish.
NMDG” Ruthenium Red (101)

 

 

 

 

 

 

 

21

Summagy of transduction mecha_nism

Some recent reviews try to form a model of olfactory transduction. The

 

general perspectives of these reviews assign second messenger systems an
exclusive role in the transduction process (examples are ref. 112, 38). These
reviews neglect to incorporate and discuss quite a few reports that do not fit this
perspective. A smaller number of authors chose not to mention these
observations at all. Table 3 contrasts the existing knowledge about the different
mechanisms. 1

Although the two pathways (CAMP and IP3) and the intracellular Ca2+ provide
a reasonable scheme, some questions are still unanswered (as shown in table
3). The concentrations of odorants needed to activate the second messenger
systems are several orders of magnitude higher than those observed in vivo or in
vitro (31 ). The data for AC activity do not correlate with the physiological data.
Bruch and Teeter (16) compared the AC stimulation in catfish by amino acids to
the potency and specificity shown by Caprio (20) in vivo. The discrepancies led
them to conclude: " transmembrane signaling mediated by cyclic nucleotide cascade in
olfactory cilia appears to depend on stimulus concentrations that produce sufficiently high levels
of receptor occupancy. Thus, altemative transduction pathways may mediate olfactory responses
to stimulus concentrations at or near electrophysiological thresholds". The odorants used in
this study among quite a few others do not stimulate the olfactory AC nor PLC.
Finally, odorants stimulated CAMP production in Xenopus melanophores at
similar concentration to those used to obtain response from olfactory neurons
AC (69). Breer and Boekoff (13) claimed that odorants that do not activate AC
induce formation of IP3. A closer examination of the data published in the
manuscript reveals low correlation between the activation of AC and PLC and

thus the relationship between formation of cAMP and lP3 is unclear. Anholt (6)

22

suggests that signal transduction at low concentration of odorants is mediated
via odorant-activated channels while prolonged exposure to the odorant
generates CAMP.

The model that suggests an exclusive role for second messenger system is
based on two major arguments: the model is similar to other transduction system
(mainly visual) and all the necessary components of the model have been
identified. Although no direct evidence for the cascade of activation of all the
components has been provided yet, quite a few pieces of indirect evidence have
been published. The major drawbacks of the model are: the extremely high
concentration of odorants required to induce response and the lack of
mechanism to support hyperpolarization. The relevance of these transduction
mechanisms to olfaction is almost unanimously accepted. The existence of
odorant-gated channels has clear evidence (139, 63). The odorant gated
channels Show response to odorants at concentrations similar to those observed

in vivo. The question of the role of the various mechanisms is still unsettled.

23

 

 

Table 3. Evidence relevant to the mechanisms of odorant transduction.
Subject Second Messenger Others
Hypothesis
Threshold to odorants. Higher than 1 (M In vivo: few 0.1 nM

In vitro: few 0.1 nM

 

The effect is unique to ORC.

cAMP is produced both in
nerve endings of the
trigeminal nerve and in
melanocytes at the same
concentrations.

No documentation of effect of
nM concentrations on other
cells.

 

Correlation between in vivo
and in vitro activity.

1. No correlation between
potency in vivo and
adenylate cyclase activity

2. No negative correlation
between CAMP production
and IP3 production.

3. Isovaleric acid which is a
primary odorant does not
induce higher levels of CAMP
nor IP3.

 

Effects of the odorant on
ORC.

1. Activation of adenylate
cyclase.

2. Activation of
phospholipase C.

3. Activation of adenylate
cyclase is correlated with
solubility in hydrophobic
media.

4. Some odorants induce
hyperpolarization but no
channel gated by second
messengers was found to
support the hyperpolarization.
5. Time course of the second
messenger production fits the
in vivo electrophysiological
response.

6. Some odorants do not
activate neither adenylate
cyclase nor phospholipase C.

Chemosensitive K+ channel
has been observed in two

species.

 

 

Components of the system
have been identified.

 

All the postulated
components have been
identified but have not been
shown to work yet.

 

Odorants induced single
channel activity in
reconstituted systems. The
receptors and Channels are
not identified biochemicaly

 

 

24
Implication of anatomical and cellular data

Some conclusions can be derived from the anatomical data and from the
features of ORCS. Observations from in vivo preparations are expected to differ
quantitatively and may be qualitatively different from those obtain in vitro.
Discrepancies exist between in vitro data as well. Two of these discrepancies
are the mechanism of transduction and the electrotonic structure of the ORC. A
model proposing second messenger mediated transduction is gaining wide
popularity but falls short of explaining the transduction of some odorants. The
ORC has a high input resistance and can generate action potentials in response
to a single opening of a ciliary ion channel. However large conductance Cl’ and
K+ channels are reported to reside in the dendritic knob.

Comparing results of different studies can Show quantitative inconsistencies.
The quantitative differences can arise from various factors that influence the
response to odorants. These differences are readily manifested in the difference
between the dynamic range of odorant concentration in vivo and in vitro. The
dynamic range of odorant concentration in vivo is 2-4 log units (45). In vitro
experiments, including in situ recordings, show a dynamic range of 1-2 log units
(39, 38). Threshold for detection of air-Dome odorants is in the range of 10-13 to
10'4 M (in humans). Similar thresholds have been measured in other vertebrates
(66). Determination of the actual concentration of odorant in aqueous solution
poses a problem since many of them have to be solublized using alcohol or
other polar organic solvents before they are introduced to the aqueous medium.

Among the factors that influence the detection of an odorant is the
amplification of the olfactory signal. Some of the factors that modulate the
stimulus-induced signal amplification are prereceptor factors while others are
neural mechanisms. One of the prereceptor factors is the olfactory binding

protein (OBP). OBP is thought to influence the odorant concentration at the

25

receptor cell membrane by providing a facilitated diffusion path for hydrophobic
odorants (95). Another factor that contributes to the amplification is the area of
the cilium membrane. Cilia increase the area of the cell's plasma membrane
considerably. Taking into account the presence of tight junctions just below the
olfactory knob, the presence of the Cilia increases the exposed membrane area
up to 100-fold. Cilia membrane has a high density of intramembrane particles,
presumably receptors. An intercellular factor that contributes to amplification is
the transduction mechanism. There is a large body of evidence for the
involvement of second messenger systems in the transduction of many of the
olfactory signals. Finally the convergence ratio of ORC to secondary cells is 100
- 1000 : 1. Under some conditions this convergence can provide amplification of
small intensity signals. Consideration of differential amplification complicates the
evaluation of the relationship between of odorant concentration and the
magnitude of the resulting neural signal.

Other quantative parameters of the system need to be considered. Affinity of
the olfactory receptors is thought to be relatively low (around 10'2 M). This is in
agreement with some suprathreshold measurements as well as chemical
modification experiments (113). Response kinetics of the olfactory system
include: initial delay, response, fast adaptation, slow adaptation and termination.
The initial delay is due to the diffusion time and a cellular delay. The slow
adaptation is believed to be in the cellular level while the fast adaptation is due
to central processing. Termination is achieved by removal of the odorant by the

flow of the mucus and possibly via metabolic removal.

Proposed theories and mechanisms of olfaction.
Olfaction has been studied from many perspectives: psychological,

physiological, pathological, environmental and more. The electro-physiological

26

studies began with electro-olfacto—grams (EOG) in the 1950's. Since then,
almost all commonly used electrophysiological techniques have been applied to
the olfactory system.

Proposed theories of olfaction can be divided to two major classes:
mechanistic vs. Classification. Classification-based theories grow out of the
notion that the olfactory system is built around the "important" (i.e., frequently
repeated ) descriptors of odors. Classification is based on verbal description,
similarity rating, etc. Amoore (3, 2) and Beets (10, 11) bridged the gap between
the verbal description and the molecular structure by classifying both odors and
odorants. Mechanistic theories proposed mechanisms by which the odor
information is transduced. In some of these theories, attempts were made to
correlate features of the odorants to perceived qualities of the induced odor.

Quite a variety of transduction mechanisms were proposed for olfaction. The
models can be divided into two broad categories: those which suggest receptor-
Iigand interaction and those which rely on some other mechanisms. More than
50 theories have been proposed in the last 100 years. Table 4 lists some of
these theories (based on ref. 19).

Advances in the study of molecular biology of olfaction suggest the existence
of proteinaeous plasma membrane receptors for odorants (ref. 78 is a recent
review). The receptors are linked to conductance changes via second
messengers and/or directly. Table 2 lists the identified ion channels in olfactory

receptor cells.

27

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 4. Proposed theories for olfaction and odorant stmcture.
Author Date General class Features

Ogle 1870 Vibrational Vibrations affects nasal pigment which gives out heat that excites
the olfactory cells.

Woker 1906 Chemical Unsaturation isthemain causeofodorbutnotessential ifsubstance
Is very volatile.

Faber 1911 Vibrational Human olfaction to material particles. Limited to insects.

Merchand 1915 Chemical Unsaturation. including trietery carbons ( CsO). Two point of
unsaturation reduces odor.

Homing 1916 Chemical Osmophore groups are important, but their relative position
determines the type of odor.

Heynlnx 1917 Vibrational Vibrations in the UV range.

Backrnan 1917 Chemical Both water and lipid solublllty are Important.

Teudt 1919 Vibrational Electronicvibration resonancaofthesensorynervesandtheodorant
molecules.

Durrans 1920 Chemical Residual affinity.

Heller 1920 Chemical Direct chemical action of the nerve endings.

Ruzicka 1920 Chemical Osmophore and osmoceptor

Tschirch 1921 Chemical Substances must be soluble in air. Loose compound with plasma of
olfactory cells.

Zwaardemaker 1922 Chemical- Odriphore. Must be volatile, lowers surface tension, lipid soluble.

Vibrational Theodfiphoredependsonthavibrationsinthemolecule.

Ungerer 8 1922 Vibrational lntennolecular vibrations within defined frequency. Effects of

Stoddard interference and resonance.

Delange 1922 Chemical Unsaturation.

Missenden 1926 Chemical Quality depends on the nature of reaction betmn odorants and
lipids. Intensity depends on number of molecules reacting.

Nicol 1926 - Function of the sinuses

Pirrone 1929 Chemical Two osmophores, one determines type of odor and the second
determines variety.

Niccolini 1933 Chemical Volatility. Solubility in the mucosa. Oxidizability.

i Krisch 1934 Vibrational Insects

Muller 1935 Physical Odorants are dipoIar. Irritata the molecular fields of the canioceptor
in the nose.

Dyson 1937 Vibrational Volatility. Lipid solubility. Raman shift between 1400 and 3500 cm'1.

 

 

28

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 4 (cont'd)
Author Data General class Features

McCordS 1939 Electrochemical Changeinbonding angleofodorant molecules In solution lnthe

Witheridga mucosa.

BeckSMIIes 1947 Vibrational IR radiationfromthemptorsisabsorbedbyhaodorants.

Baradl a Bourna 1951 Enzyme Inhibition of enzyme action by odorants.

Hainer 1953 Information Thirty levels of intensity; 24 Idnds of primary odor.

Wright 1954 Vibrational Raman shift lower than 900 - 1ooo om'1 .

Davies 1954 Physico- Puncturingofolfactorycell membraneandexchangeoma‘ and K‘.
Chemical

Moncrieff 1961 Physical Volatility, adsorbability and customary absence from olfactory region.

Amoore 1962 Stereochemical Primary odorants. Shape determines odor quality.

Beets 1964 Stereochemical - Part of the odorant molecule presents an odor active ”profile".
functional
9'0“”

Polak 1973 Stereochemical - Multiple profile - multiple receptor site.
functional
9'0““

Kurlhara 1987 Chemical Adsorption of odorants to the membrane causes change In

mefnbranepotentialandfklidity.Thellpldcomposltiondatefmlnes
specificity.

 

 

 

 

 

29

Odorants were found to activate adenylate cyclase (92) as well as
phospholipase C (12) in ORCS. Ion Channels gated by CAMP (85) and by IP3
(101) were identified. These findings are consistent with a model that suggests
involvement of CAMP, IP3 and Ca2+; in the transduction of olfactory signals. No
explicit model has been proposed for the coding of odorants but analogy was
drawn to the immune system. Recent identification of a multigene family that
encodes proteins similar to G protein-coupled receptors in other systems might
provide the missing link (17, 18).

Non-receptor theories of transduction proposed that the odorants act on the
ORC's plasma membrane without the mediation of a specific membrane protein.
Mechanisms such as energy transfer through resonance to cytoplasmic enzyme
inhibition were proposed. Davies's puncturing theory (26) and Kurihara's
adsorption theory (88, 34) are the only two of the non-receptor theories that
have not been rejected completely. The puncturing theory proposes that
hydrophobic odorants puncture the plasma membrane of the ORC. The puncture
causes a local short circuit and thus depolarizes the membrane. The larger the
molecule is, the better a stimulant it is. Quality of odorants depends on the
relative rates of diffusion through the mucus and the rate of "healing" of the
punctured membrane. Since the physicochemical basis of the theory is
reasonable and correlation was found between olfactory threshold and an index
of the cross sectional area of many odorants, the theory has not been
categorically rejected. Kurihara and co-workers found that various odorants
depolarized plain lipid Iiposomes. The lipid composition of the liposomes had
changed the odorant threshold concentration. In support of this mechanism,
Enomoto et al. (34) point to the observations of odorant activation of non-
olfactory cells. Adenylate cyclase of melanophores was shown to be activated by

odorants. Turtle's trigeminal nerve cells were shown to be activated by odorants.

30

To account for different specificity of ORCS, the model postulates that the lipid
composition of ORCS is not uniform.

Although all but the most current theories have faded away, some of the old
theories are worthy of a reexamination. For example, vibrational theories grew
out of analogy to vision and hearing. One of these theories was successful in
predicting human olfactory qualities based on far IR spectra of the odorants. Far
IR spectra describe sub-structures of many organic compounds. The absorption
of many bonds common in organic compounds is at this range and thus
information both on the bond and on the stress on the bond can be obtained.
Enzyme-based theories proposed allosteric effects of odorants on receptor cell
enzymes. Adenylate cyclase activity has be shown to be influenced by changes
in membrane fluidity. Chemical theories can be divided into those which
proposed Chemical reactivity and those which proposed the physicochemical
properties (2). The Chemical theories stressing the importance of the
physicochemical features of odorants have evolved to the
stereochemical/functional groups model that is widely accepted today.

Several models were proposed for odor coding and discrimination (41, 107, 1,
15). In recent years these models employed the latest ideas from mathematics
and computer science. All the models agree on two important points: (a) a large
majority of olfactory receptor cells are not specific to a single odorant and (b)
neurons become more specific to an odorant with the level of the processing
stage (124). None of the models has tried to tackle the question of the variety of
stimulants to which an ORC responds. Detailed discussion of the models is
beyond the scope of this review.

To summarize, the current widespread view is the that transduction depends
on membrane receptors. The description of the receptors is still enigmatic. Many

of the components involved in transduction have been identified. A complete

31

model has not been formulated yet. Non-receptor theories have been rejected
(generally) but some of these studies contain valuable information about

possible Classification of odorant molecules and their features.

Structure-activity relationship

With all the knowledge that has accumulated about the structure of the
olfactory system, about the molecular biology of ORC and about the
mechanisms of olfactory signal transduction one aspect remains obscure. What
exactly is the stimulus? In other words, what is the relationship between the
structure of the odorant and the olfactory signal? This question has been named
"Structure-Activity Relationship" (SAR). Beets wrote: "It is the purpose of research on
SAR in olfaction to identify the major types of olfactory information and to find out which
structural parameters of the odorant molecules are involved in the generation of each type"
(10).

Most of the data about SAR in olfaction comes from psychophysical
experiments complemented by a Chemical analysis of the odorants. Data from
physiological experiments were obtained in a few cases.

The most recent model that focuses on SAR is the "multiple-functional groups
multiple-profiles" model. This model is based on the ideas that were introduced
in the "stereochemical theory of olfaction". The evolution of the idea that the
odorant molecule shape and the existence of a set of specific receptors are the
key to olfaction dates back to the 1920's. Incorporation of data about a variety of
features of odorants and their relationship to the perceived odor resulted in the
current notion of receptors to some physico—chemical structures of the odorant

molecules.

32

Amoore's stereochemical theory

Amoore and co-workers asked a panel of judges to rate the similarity of over
100 compounds to 7 standard odorants (scale of 0 to 8). The standard odorants
were in their view representatives of the "primary odors". The ratings were
compiled to a table of average similarities. Out of the 7 odorants, only 5 were
found to yield statistically valid results. In parallel, three silhouettes of each of
the compounds were prepared. Each of the compounds was sterically modeled.
The top, front and right silhouettes of the models were photographed. The
silhouettes were compared to the silhouettes of the standard odorants and the
results were compiled into a similarity table analogous to that of the similarity
ratings. Statistical correlation (by calculating least squares based linear
regression) was calculated between the two tables. Amoore found statistical
correlation between the similarities of shape to the similarities of odor for each of
the five standards. The results of this analysis led Amoore to suggest that odor
quality (olfaction modality) is related to molecular shape, i.e., the
"stereochemical theory of olfaction" (3). Seven primary odors were proposed:
(a) ethereal (1 ,2-dichloroethane), (b) camphoraceous (1 ,8-Cineole), (c) musky
(15-hydroxypentadecanoic acid lectone), (d) floral (d,l-B-phenylethylmethylethyl
carbinol), (e) minty (d,I-menthone), (f) pungent (formic acid) and (g) putrid
(dimethyl disulfide). The theory was challenged but the general principle of
receptors for structural elements of the odorant molecule remains accepted.
§flific anosmia

A more reliable identification of the molecular structures that are detected by
the olfactory system was attempted by studying specific anosmia. Anosmia is the
inability to smell. Some humans cannot smell specific odors. These people are
said to have specific anosmia. Specific anosmia is of a particular interest to the

study of olfaction because of the analogy to color blindness. Anosmia was

33

suggested to be the result of specific deficiency at the receptor level. If this
hypothesis is correct, an anosmic nose reflects the inability to detect a particular
stereochemical structure.

A group of odorants is evaluated by anosmic testers. This evaluation is
compared to that done by normal testers. The evaluation is for the threshold for
detection. The chemical structures of the odorants are analyzed in reference to
the anosmic defect factor (reflects the ratio between the detection thresholds by
normal vs. anosmic testers). One of the odorants yields a particularly high factor.
This odorant was termed by Amoore (2): "a primary odorant". An example of
such a study is the aliphatic aldehydes (in particular iso-aldehydes). The highest
anosmic defect factor is for butyraldehyde. The tested odorants differ in the
number of carbons in the aliphatic Chain. Butyraldehyde is the smallest of this
group. The threshold of detection among anosmic testers decreased as the
length of the aliphatic Chain increased suggesting that the rest of the molecule
(the iso tail and/or the chain) is detected by another kind of receptor. No account
was given to the qualities of odor the testers reported (butyraldehyde has a
malty odor). Based on studies of this kind Amoore reports 8 primary odorants
(different than primary odors) : (a) isovaleric acid (sweaty), (b) l-pyrroline
(sperrnous), (c) trimethylamine (fishy), (d) isobutyralde hyde (malty), (e) 5a-
androst-16-en-3—one (urinous), (f) co-pentadecalactone (musky), (g) l-carvone
(minty) and (h) 1,8-cineole (camphor). The number of primary odors defined
using specific anosmia depends on finding people that have this defect and is
expected to grow substantially. Specific anosmia provides a method to identify
molecular structures that normally activate olfactory receptors. Although it is
reasonable to assume that the basis for specific anosmia is at the receptor level,
no direct test has been applied to validate this assumption. AS a footnote it is

interesting to note that the type smells of primary odors is of chemicals or plants

34

while the odors induced by the primary odorants are mainly (except carvone)
animal related odors. From an evolutionary point of view this observation seems
to be reasonable.

Beet's functional groups

While the stereochemical theory of olfaction and specific anosmia have tried
to find some common features in the odorants, these approaches do not attempt
to explain a general principle. Grouping a large number of odors reveals low
statistical validity. Analyzing the hierarchical grouping of as few as 74
descriptors of commonly used odorants by hierarchical Classification of a
similarity matrix led Chastrette et al. to write: "This study shows that the notes are
generally independent, with no strict hierarchy among them, and rules out the existence of
primary odors." (23). This observation supports the notion that it is not the
perceived odor that represents the stimulus.

Beets (10, 11) and Polak (96) took a different approach to analyzing SAR.
The foundation for the analysis and the model that followed is the informational
structure of olfaction. It is a known phenomenon that almost all odorants exhibit
to a human tester more then one quality (modality, facet). However, many of the
odorants have a predominant quality and one or more minor qualities. What are
then the features that encode a quality?

Beets formalized the discussion of qualities by suggesting that if an odorant
exhibits the descriptor A then all the odorants S1 to Sn that belong to the same
Class will have the same quality (possibly among others). This formalization is
supported by data from people with specific anosmia. The data from specific
anosmia shows that people that cannot detect a particular odor quality exhibit
high detection threshold for compounds with a particular molecular structure.

From this argument a suggestion can be made that descriptor A represents: "the

35

same principle in the odors of this class of stimulants and that principle is a homogeneous,
indivisible informational component of these odors".

Since descriptor A is the end product of the olfactory process, Beets
suggested that descriptor A is the terminal derivative of an informational
modality. In other words the derivative of a homogeneous informational
component of the peripheral information pattern that results from the stimulation
of the olfactory receptors by any odorants from the Class S1 - Sn. The concept of
informational modality in olfaction can be defined as: the common informational
element in peripheral information patterns generated by interaction of an olfactory epithelium
with all members of a set of stimulants (odorants) producing similar odor impressions.
Following these definitions, the stimulation of the OE by a set of odorants that
elicit the same odor (possibly among others) is expected to produce an activity
pattern that contains a common sub-pattem.

Beets formulated two postulates: (a) there is a limited number of discrete
informational modalities and all members of the same species depend upon the
same modalities and (b) stimulation of the epithelium with a pure odorant
(structurally, configurationally and Chirally homogeneous) produces a complex
pattern of information that contains a number of modalities at various intensities.

The first postulate is supported by data from specific anosmia and from the
fact that descriptors among a group of testers are similar. The second postulate
requires an explanation. The initial event in odorant detection is the interaction
between the odorant molecule and a receptor. The molecule, being a 3-D non
uniform body, can be presented to the receptor in various angles. Thus the same
molecule even without conformational changes can present multiple profiles to
the olfactory receptor sites. Along with the multiple angles, flexible molecules
can assume many conformations (specially when interacting with a receptor).

The multiple configurations and the different angles yield quite a few profiles of

36

the same odorant molecule. The second postulate defines a sharp distinction
between the concepts of odors and informational modality in olfaction. Testers
with specific anosmia to the primary quality (modality) report other qualities.

Odors can be Classified into three major groups: those that have a single
strongly predominate modality, those that have several modalities that stand out
and those that have many modalities all of which are in low intensity. Primary
odors are the extreme of the first group. Single modality can arise from a unique
sterically accessible functional group (96). Functional group in this context is a
part of the molecule that interacts with a receptor and leads to activation of the
receptor.
Mme in support of the fgn_ctional groups model

The notion of independence of the profile-functional groups receives
additional support from physiological experiments. Among these experiments
are: the response to mixtures of odorants, chemical modification of the receptor
site and single Channel recordings.
Mixtures

Studies of mixtures of odorants are done by comparing the responses to each
of the odorants presented separately versus the response to a mixture of the
odorants. Three phenomena are observed: (a) suppression, (b) synergism and
(c) a different odor. Suppression refers to a response to the mixture smaller than
the sum of responses to each of the odorants. Synergism is the opposite, the
response to the mixture is larger then the sum of responses to each of the
odorants. The interesting observation among the three is the third: a different
odor (45, 96). Using this phenomenon is a common practice among perfumers
and flavorists. This means that a single compound can be replaced by a mixture
of two or more odorants (in particular proportions). For example cedrol can be

replaced by a mixture of santalol and camphor (96).

37

Chemical modification

Chemical modification of a receptor site is done by applying an agent
(compound or enzyme) that covalently modifies the receptor protein and hence
inactivates it. Examples are the thiol specific reagent mersalyl and
lacoperoxidase (catalyze iodination) (113, 81). The epithelium is exposed to the
modifier agent in the presence of an odorant. After the treatment the agent and
the odorant are washed from the epithelium. Responses of the epithelium to a
series of odorants are then measured. These responses are compared to
responses of DE that was exposed to the modifying agent without the odorant
present at the time of the exposure. The comparison shows that an odorant can
"protect" its receptor(s) from modification. It also suggests that there is not a
single receptor for each odorant but rather a number of receptors will respond to
unrelated odorants as long as the odorants have some structural similarities.
Summam of structure-activity relationship

On the basis of the relationship between odor and chemical structure of
odorants and the data about specific anosmia, a model was formulated. Because
of the all or none nature of specific anosmia, the deficiency is thought to be due
to lack of a particular receptor. Extension of this notion is the postulation of a
limited number of receptors to functional groups-profiles. A profile together with
a functional group(s) is defined as a peripheral informational modality.

The more modern studies of SAR concentrate on the receptor cells and the
receptor sites. Beet's model is still the working model in SAR studies. Data from
experiments with mixtures of odorants and from chemical degradation of OE
receptors support the presence of receptors to sub-structures of odorant

molecules.

38
Structure of odorants

With respect to SAR it is appropriate to look at the odorant molecule features.
What is known about specific odorants and their SAR? The first notion of
structure-activity relationship (SAR) in olfaction dates to the Roman poet
Lucretius who wrote around 55 BC: ”So that you may easily see that things which are
able to affect the senses pleasantly, consist of smooth and round elements; while all those on the
other hand which are found to be bitter and harsh, are held in connection by particles that are
more hooked and for this reason are wont to tear open passages into our senses" (19). The
nature of the olfactory stimulus has gone through a long and winding road to
come back to the same notion during the early 1960's. Most of the modern
research aimed at the structure of odorants involves detailed studies of SAR in a
particular class of compounds.

General features of odorants ‘

A major enigma in the research of olfaction was and still is the qualitative
aspects of odors. Because of the large number of odoriferous compounds it is
difficult to generalize features of odorants. However, few things can be stated:
(a) the highest molecular weight of an odorant found so far is 294, (b) most
odorants are uncharged and do not undergo a metabolic transformation and (c)
most odorants contain a weak polar region and a strong hydrophobic region
(90). The weak polar region was termed the "osmophore". It appears that the
immediate molecular environment of the osmophore plays an important role.
Osmophore notion was expanded to a description of the stereochemical features
that define an odor (91 ).

Many attempts were made to define those parameters of the odorant
molecules that have relevance to olfaction. Odor Classification attempts vary
from 146 odors to 4 odor Classes. The notion of primary odors has Changed to

primary profile-functional groups. Classification however, remains a powerful

39

technique. Edwards and JurS (33) correlated odor intensity to structural
properties of the odorants. This study identified the molecular properties that
make an odorant potent. All the parameters that were found to be important fit
Beet's model. The five parameters are: the log of the molecular weight, the
partial charge on the most negative atom in the molecule, the quantum
mechanical polarity parameter, the average distance sum connectivity and a
measure of the degree of unsaturation in the compound. Higher molecular
weight usually corresponds with a larger, more complicated molecule and thus a
molecule that contains more profiles-functional groups. Since the odorant
molecule interacts non-covalently with the receptor, a larger partial charge
should contribute to the strength of the interaction. Polarity of the odorant
restricts the number of orientations of the odorant molecule when it approaches
the charged receptor. The observation of increased threshold with increased
polarity fits this prediction. Branching (average distance sum connectivity) was
found to decrease the threshold. Branching increases the complexity of the
molecule and thus the number of possible profiles. Increased number of profiles
is expected to reduce the threshold. Unsaturation (presence of it electrons)
decreases the threshold. The effect of the presence of it electrons is similar to
partial charge in regard to electrostatic interactions.
Examples of known odorant SAR

Few studies have defined the parameters of the odorant that appear to
determine its olfactory modality. Understanding these parameters can assist in
predicting the nature of the binding site in the receptor.

An extensive treatment of SAR in ambergis odorants has been done by Ohloff
(89, 90, 91). Ambergis odorants induce a few types of odors: wet mossy, strong
tobacco, balsamic, musk tonality, seaweed and fecal. The basic structure of the

odoriferous compounds is a di or tricyclic alcohol. By examining the structure of

40
the various compounds Ohloff was able to describe many of the salient
parameters like electronic properties, hydrophobicity, lipophilicity and steric
structure. A model based on the steric arrangement of two atoms and a
hydrophobic region of the odorant molecule was proposed (90). Ohloff proposed
some modification and additions to Beet's theory. The major differences are: that
structural elements and not functional groups should be regarded as
osmophores and that hydrophobic interactions are the major driving force for
substrate binding (at least for this group of compounds). SAR of fragrances is
developed for musk odorants as well.

Caprio (20, 21) studied the receptor specificity of the catfish olfactory system
to animo acids. By measuring threshold Concentrations and cross-adaptation,
four subtypes of receptors could be defined. Caprio was also able to provide
insight into the specificity of the receptors by using substituted amino acids and
comparing the response to them. Kang and Caprio (60) have recently showed
that the response to a mixture of odorants could be predicted based on
information about two factors: (a) the responses to individual components and
(b) the relative independence of the receptive sites for each of the components.
They also note that a synergistic response might be impeded by the limitation of
the output of the ORC. These authors use an index termed: "mixture
discrimination index" (Equation 1). The different components are introduced in
concentrations that produce the same magnitude of response (equipotent). The
response to the mixture is then normalized by the average responses to the
same concentrations of the component odorants when introduced individually.
The mixture discrimination index (MDI) is useful in quantifying the degree of
independence of the responses to each of the components. MDI equals 1 if the
components bind to the same receptor, greater than 1 if the components bind to

different receptors and less than 1 if there is mixture suppression.

41

Equation 1
MDI _=. R"

R; = Response to the individual components.

Where:

Rm = Response to the mixture.

n = Number of the components.

The odorants used in this study.

As stated in the beginning of the introduction, the study was aimed to provide
evidence for the existence of more than one osmophore in an odorant. The
theory equates osmophores to molecular substructures. Thus the goal can be
stated as the search for evidence of a relationship of an odorant with more than
one receptor. The evidence consists of conductance Changes that point to
binding of an odorant to more than one receptor. The receptors have not been
identified yet. Thus, the system has to be probed by two odorants that share one
osmophore but differ by at least one.

The perspective of this study is SAR. By using diethylsulfide (DES) and two
structurally related molecules an insight on the receptor for one functional group
can be obtained at the most crucial level: the receptor level. The major aim was
however to Show that a single odorant molecule expresses two osmophores. The
analysis of responses to DES and I—carvone at the single Channel level is an
example of SAR perspective in receptor level studies. In a reconstitution study
(139) the same Channel was activated by both DES (rotten egg odor) and I-
carvone (Spearmint oil odor).

A Close look at carvone reveals a part of the molecule that looks similar to

DES. The concentration needed to activate the channel with carvone was only

42

slightly higher than that of DES. As depicted in F lgure 1, part of the carvone
molecule is similar to the DES molecule. Both the sulfur of DES and the oxygen
of carvone's are partially charged and the locations of two of the hydrogens
(marked by arrows) are similar in both molecules. The fact that the only similarity
in molecular structure between the odorants is a part of the carvone molecule

supports the concept that receptors react with parts of an odorant molecule

 

 

 

 

 

(osmophores).
A
B
\
Figure 1. Structures of diethylsulfide (A) and l-carvone (B). Both are

viewed from a similar plane (carvone's ring plane)

Three odorants were used in this study (Figure 2): diethylsulfide (DES),
diethanolsulfide (thiodiglycol) (DOS) and thiophene (THP). Thiophene has been
shown to induce hyperpolarization of ORCS (64). Diethylsulfide increases the
open probability of a K+ channel (139, 134). DES activates a receptor that

increases K+ conductance. Hypothetically, so do thiophene and diethanolsulfide.

43

THP shares the functional group -CH2-S-CH2- but differs considerably in its
physicochemical properties (thiophene is a rigid ring molecule but smells similar
to DES). DOS has the same -CH2-S-CH2- group but has distinctly different odor
(a sweat odor). DOS is essentially DES with -OH groups at the ends. Because
of the similarity and the difference, DOS is expected to activate some other path
along with the same K+ conductance. This path could be either another K+

conductance or conductance of another ion.

 

 

 

 

 

Figure 2. The three odorants used in this study. (A) - Thiophene, (B) -
Diethanolsulfide and (C) - Diethylsulfide.

The three compounds are thiols with two alkyl or modified alkyl groups. The
common denominator of the three compounds is the sulfide group. Diethylsulfide
is a sulfur atom connected to two alkyl groups. The sulfur is bound to two
carbons (sp2) and thus left with two 1) electron. The bound alkyl groups form
bulky hydrophobic regions. Diethanolsulfide is a derivative of DES. The last
carbon of both alkyl groups is replaced by a hydroxyl.

44

The electronic structure of oxygen is similar to that of sulfur (both in group
Vla). Oxygen that has four sp3 hybrid orbitals when in a hydroxyl group contains
large partial Charge. As a 4 carbon, hetrocyclic and unsaturated compound,
thiophene has 7 sp2 type bonds (C-C, C-H and C-S), two it orbitals contributing
4 it and two 11 electrons from the sulfur. The two 11 electrons provide partial
charge that is dissipated to some extent by the resonative hybrid structure (77,
50, 56).

lntramolecular distances and angles were calculated using a molecular
modeling program (Alchemy ll, Tripos Associates Inc). The program calculates
the most stable configuration of a molecule by minimizing the intramolecular
forces. The distance between the two most remote hydrogens was taken as size.
The Sizes are as follows: DES - 6.263A, DOS - 6.960A and THP - 4.578A. The
C-S-C bond angles are: 985°, 984° and 863° respectively. The distances to the
first hydrogen are: 2.416A, 2.424A and 2.578A respectively. From the
geometrical data it appears that THP is a more compact molecule with the
smallest steric hindrance for the sulfur. THP is also a rigid molecule with a single

conformation.

Hypothesis

The underlying hypothesis of this study is that the coding of on odorant
stimulus is by the activity of many cells each of which contains receptors for
different functional groups - profiles (osmophores). The integration process is
done in the olfactory bulbs and higher centers. It is possible for two osmophores
to activate the same ionic conductance and still maintain specific information
about the odorant as long as the outcome of some integration process such as

the activity pattern of the sensory cells, is different. This will occur if the

45

distribution of the receptor types is not uniform or if some other receptor is
activated in the sensory cell by another osmophore of the same odorant.

The working hypothesis of this study is that DES and DOS present a common
osmophore but differ from each other by at least one additional osmophore
presented by DOS. The difference between the response to DES and DOS might
be in the activity of a single cell type or the pattern of active receptor cells even
if both increase conductance to K+ alone. Because of its chemical structure THP

was expected to resemble DES.

Reconstitution

The aim of any experimental system is to eliminate as many uncontrolled
factors that influence the outcome of the experiment. The study of biological
membranes and their embedded proteins can be greatly influenced by a large
number of processes that occur naturally as well as by the experimental
procedure. The isolation of membrane fragments or components and then
positioning them in a highly controlled system enabled investigators to perform
experiments which were otherwise impossible. Thus, the main objective of
reconstitution is the incorporation of isolated membrane components into a well-
defined experimental system. Membrane proteins were reconstituted in
lyposomes as well as cell membrane. The reconstitution in lyposomes is
designed mainly to provide a system to investigate ion fluxes. Expression of
genomic material of membrane proteins facilitated both biochemical and
electrophysiological studies. The most common reconstitution of ion Channels is
done by incorporation into planar BLM or into a BLM formed at the tip of a glass
pipette.

A large variety of membrane proteins were reconstituted. For the purpose of

this introduction, the pertinent proteins are ion Channels and especially ion

46

channels from the ORCS. Many types of ion channels have been reconstituted,
most notably the nicotinic acetylcholine receptor (35, 109, 86 and many others).
The proteins that constitute the acetylcholine receptor were isolated, identified
biochemically, Cloned and inserted into liposomes and later into BLMS. The
acetylcholine receptor is an example of the utilization of reconstitution to its
fullest as different mutated Clones provided information on the relationship
between the structure of the proteins and their function. Most reconstituted ion
channels were not studied to the same extent as the nicotinic acetylcholine
receptor. The majority of the reconstitution studies used biochemically isolated
channels that were incorporated into a BLM rather than Cloned proteins.

While the plasma membrane Channels are accessible to probes used in
electrophysiological studies, the ion channels of the cell organelles cannot be
approached. The only possible way to study the properties of the Channels of
internal organelles is by reconstitution. A sarcoplasmic K+ selective channel (82)
and a Ca2+ Channel (106, 120, 130, 131) were reconstituted. Brain microsomal
Ca2+ Channel (132) was reconstituted as well. Protein-conducting Channels from
endoplasmic reticulum were reconstituted (118, 117).

Similar to the membranes of internal organelles, some plasma membranes
are not readily accessible. The Cilia of the ORC are small in diameter and thus
are not feasible for patch-Clamp studies. Some ciliary ion channels were
reconstituted: two different odorant-gated Channels (139, 63), a cyclic
nucleotide-activated Channel (1628) and an lP3-gated Ca2+ Channel (101, 102).

Reconstitution provides another important tool for the study of ion channels: it
isolates the membrane proteins from the influence of the cytoplasmic regulatory
machinery. Many ion channels were found to be regulated by cytoplasmic
enzymes. During the membrane protein isolation procedure the relationship

between the individual protein and the cell it was a part of are severed. The

47

isolated membrane fragments or proteins are a mixture of parts of many cells.
This point is particularly important in the study of receptors. It has been shown
that some topographical arrangement of receptors exists in the olfactory
epithelia (76). The deciliation procedure that fragments the plasma membrane of
ORC cilia into small pieces, results in a mixture of all the receptors. We selected
the reconstitution method to study the responses to structurally related odorants
because of the three features described above: (a) the ability to voltage-Clamp
the ion channels, (b) the isolation from cytoplasmic influence and (C) the

independence from the specificity of the ORC.

Objectives of the study

The intention of this study was to provide evidence in support of the multiple
profile - multiple functional group model. This model describes how the olfactory
receptors recognize odor stimuli.

The study was designed to determine if DOS presents two osmophores: one
that is similar to the osmophore presented by DES and an additional osmophore.
Does DOS induce a conductance Change independent from the one induced by
DES and thus mediated by a different receptor? The second objective was to
determine whether DOS induces a conductance Change via the same receptor

that binds DES.

METHODS

General

The general procedure used in this study was the voltage-clamp of
reconstituted olfactory cilia membrane in planar bilayer lipid membrane (BLM).
Olfactory cilia membrane was processed to form vesicles. A BLM was formed in
an orifice through a Teflon® partition. The vesicles were then induced to
incorporate into the BLM. The current through the modified BLM was measured
under a voltage clamp. After the incorporation of olfactory Cilia membrane (OCM)
fragments into the BLM, different chemical agents were added to the aqueous
solution and their effects were determined.

Experiments to determine mixture discrimination index (MDI) were performed.
EOG in response to various concentrations of the odorants or their mixtures was

recorded in vivo. The odorants were presented in liquid phase.

Purification of membrane vesicles

Olfactory Cilia were isolated from bullfrogs (Rana catesbeiana) obtained from
commercial suppliers. Animals were 5" to 7" long and weighed from 2009m to
5009m. Geographical or seasonal differences were not observed. The frogs
were anesthetized using 0.1% (w/v) tricaine methane sulfonate and double
pithed. The nose was quickly removed and placed on an ice-cold glass plate.

Both the ventral and the dorsal epithelia'were carefully removed from the

48

49

supporting structures. Care was taken to exclude nerves and connective tissue.
The epithelia were then placed in the first buffer of the deciliation procedure.

Two methods for isolation of ciliary membrane fragments were used. The first
method is known as "ethanol-calcium shock". The procedure was adopted from
Chen and Lancet (24) and Linck (72). The epithelia were placed in 10 times their
volume of an ice-cold amphibian Ringer solution (1.1 mM CaCl2, 1.9 mM KCI,
2.4 mM NaHCO3 and 111 mM NaCl ) containing 0.1 mM EDTA. The tissue was
washed twice by replacing the Ringer-EDTA solution. This step removes the
mucus that lines the mucosa. The epithelia were then placed in a solution of
10% (v/v) ethanol (EtOH), 0.1M NaCl, 2mM EDTA, 30 mM HEPES, pH 7.0 at
room temperature (100 ml solution per 10 gm tissue). The tissue was allowed to
equilibrate for 2 minutes while stirring gently using a magnetic stirrer. CaClz
solution ( 1M) was then added rapidly to a final concentration of 10mM CaCl2 .
After 5 minutes at room temperature the suspension was filtered through four
layers of gauze and centrifuged at 1,5009 for 5 minutes. The supernatant was
centrifuged at 10,0009 for 5 minutes and the pellet was resuspended in a
HEPES based buffer. This pellet is an enriched cilia membrane fragments
preparation. It appears that this preparation is highly pure although some
contamination from cytoplasmic organelles has been reported (103).

The second method used in this study was based in Boyle et al. ( 12). The
dissected epithelia were placed in ice-cold 20 mM TRIS-HCI, 0.3M Sucrose, pH
7.4 (buffer A). The mucus was removed by washing the tissue in buffer A + 5 mM
EDTA. The EDTA solution was replaced 3 times at intervals of 5 minutes.
Deciliation was done in buffer A + 10mM CaCl2 and 10 ugram/ml Ieupeptin. The
test tube with the epithelia and buffer was mechanically agitated using a vortex 4
times for 10 seconds (at setting 6) at intervals of 5 minutes. About 1.5 ml

deciliation buffer solution were used for one gram of wet tissue. The tissue was

50

kept on ice. The suspension was then centrifuged for 10 minutes at 8009. The
supernatant was centrifuged at 28,0009 for 30 minutes. The pellet was
resuspended in a HEPES buffer containing the electrolytes used in the
experiment and 0.4 M glycerol. The resulting pellet was termed by Boyle et al.:
"initial ciliary pellet". This preparation contains some pigment but further
purification reduces the amount of protein only to a small extent, indicating that
most of the initial ciliary pellet was relatively pure (12).

All centrifugations were done in a Sorvall RC-5 centrifuge using a SS-34
rotor. Protein amounts were assayed according to the modified Lowry procedure
(modified by Peterson). The ciliary membrane fragments were resuspended in a
HEPES based buffer which is the electrolyte solution that was used in the
following reconstitution experiment, and an appropriate concentration of
glycerol.

Averaged results from the experiments comparing the two methods of
isolation are shown in Figure 3. There is no significant difference between the
mass of wet tissue obtained per bullfrog in the experiments. The amount of
protein was about twice when the isolation procedure that follows Boyle et al.
(mechanical agitation) was used. The ratio between the two yields is consistent
with the literature (12). The procedure pf Boyle et al. was used in most of this

study.

51

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i . l . . . . mmmm
as: aaaaa a mmmmm a so; www.mm
mmmm

52
BLM

Methods of constructing a setup for BLM are described in detail by Tien (125),
Vodyanoy (137) and others. BLM was formed in a round aperture in a
hydrophobic partition between two compartments that contain aqueous
solutions. Many factors are important for the stability of this extremely thin
structure. Two crucial factors are the smoothness of the surface of the orifice
wall and the composition of the forming solution. The forming solution is a
solution of amphipathic molecules in an appropriate solvent. Usually the
amphipathic compounds are lipids and the solvent is an alkane. The "painting"
technique was used to form the membrane. A minute quantity of the forming
solution was spread on the brim of the orifice and then dragged over it with a
fine brush. It is extremely important to prime the orifice before it comes in contact
with the aqueous solution. Priming was performed by depositing a small amount
of the forming solution on the side of the orifice and then drying under argon.
Priming prevents the formation of gaps between the BLM and the partition. The
formation of the BLM was monitored both visually and electrically. Since the
membrane thickness is in the same order of magnitude as visual lights wave
lengths, interference colors appear. The membrane starts as a thick layer, goes
through a thin layer stage and then thins to a bilayer. When the BLM is
completely thinned, the thickness of the BLM is such that destructive
interference causes a black appearance. In parallel to the visual inspection,
electrical monitoring of the membrane thickness was done by measuring the
capacitance. Capacitance is inversely proportional to the thickness of the
dielectric element and proportional to its area. Since the area of the membrane
remains relatively constant and assuming that the dielectric constant does not
change considerably, the Change of the capacitance is an indicator of the BLMS

thickness.

53

Reconstitution in a BLM requires a lipid base. Two forming solutions were
tested and used. A 2.5% soy bean total phosphatide extract (SBTPE) in n-
decane solution and a 4% SBTPE in n-decane solution. The acetone extract of
SBTPE was termed AE. BLMs from 2.5% SBTPE in n—decane had resistances in
the range of 0.5 to 3G!) (3.9 to 23 MO - cm’) with an average of about 1G0 (7.9
M!) - cm”). The membranes were stable if left undisturbed but could not
withstand excessive manipulation. To increase stability and facilitate fusion of
membrane vesicles the soy bean total phosphatide extract was purified using the
method of acetone extraction. In principle this procedure extracts the polar lipid
from the natural mixture. This is done by partitioning the mixture in
ether/acetone. The polar lipids precipitate in the non-polar solvent and are
further processed (for details see Appendix A). The composition of SBTPE is:
45% lecithin, 20% phosphatidylethanolamine (PE), 20% phosphatidylserine
(PS), 10% inositol containing phospholipids, 5% sterols and glycolipids. A
forming solution was made by drying the appropriate amount of lipids under
argon or nitrogen and then dissolving it in n-decane. A 4% solution gave the
best balance between stability and thinning. The resistance of BLMS made from
this solution was in the range of 10 to 50 GO (78.5 to 395 MO - cm’), with an
average of 25 GO (200 MO - cm”). The BLMS were stable for a few hours or
more and could withstand manipulation. The specific capacitance of BLMS made
of either of the forming solutions was similar (about 0.4 uF/crn’). Argon was
bubbled in both the forming solution and the electrolyte solution for 30 minutes.

Introducing ciliary membrane fragments into the BLM is the final stage before
the actual experiment is performed. The ciliary membrane fragments (when
suspended in electrolyte solution and sonicated) form vesicles. The vesicles are

then induced to fuse with the BLM.

54

Although the theory of fusion of membranes is far from complete, several
principles have been established. The proposed model suggests that the fusion
of membranes happens in few stages (Figure 4). Vesicles get attached to the
planar membrane. If the surface tension of the attached vesicle's membrane is
high enough compared to the BLM, the vesicle opens and flattens. Since the
edges of the opening in the vesicle's membrane are attached to the BLM it
becomes part of it (25, 54). The required surface tension is obtained by osmotic
pressure. The system was arranged to have an osmotic gradient across the BLM
as well as across the vesicle's membrane (25, 87, 118, 117). The vesicles
contain 0.4 M glycerol. Glycerol was added to the as side of the BLM (defined
as the side the vesicles are added to) to. a final concentration of 0.35 M. Care
was taken not to have an excessive osmotic gradient on the vesicle membrane
since it causes lysis of the vesicles.

The rate of attachment of vesicles to a BLM depends on the concentration of
the vesicles in Close proximity to the BLM and on the concentration of divalent
cations (25). The electrolyte solution used contained 8mM calcium acetate which
results in a concentration of about 2.5 mM Ca2+. The vesicle suspension was
injected (1-2 pl) using a micropipette located about 0.1 mm from the BLM (118,
117). This was done in order to increase the concentration of the vesicles at the
face of the BLM.

The BLM was held at a potential of +30 to +40 mV (in reference to the trans
side). The current was monitored continuously. A typical sharp increase of
current signals the fusion of a vesicle (Figure 5). If the capacitance did not
change significantly the event was considered as a fusion. Glycerol was then
added to the trans side to eliminate the osmotic gradient and thus prevent
additional fusion events. The current was then recorded under different

potentials. Many times current fluctuations typical of ion Channel activity were

55

observed (Figure 5). In order to increase the probability of incorporation of the
appropriate receptors, fusion was not halted until conductance of around 0.5 to 1
n6 is obtained. This was found to be a reasonably stable baseline conductance.

Table 2 lists the ion Channels found in ORCS. K+ Channel conductances
range from around 35 pS to 190 p8. Cl' Channel conductances range from 130
p8 to 210 p8. These conductances allow the estimation of the number of ion
channels that were incorporated into the BLM. It is reasonable to assume that
the Cl' Channels and the K+ Channels are mostly open at the range of potentials
that the current were measured (i 40 mV). Thus in an experiment where a
conductance of 1 n8 was measured after fusion, 5 to 28 channels were
incorporated.

Figure 6 shows a scheme of the measurement system. This is a two electrode
configuration. One of the Chambers (Cis side) is grounded. Cis Side was chosen
in order to reduce noise during manipulations like vesicle injection. Both
electrodes were AglAgCl connected via an agar KCl bridge (4% agar, 0.5 M
KCI). The system includes a high input impedance electrometer (Keithley 614),
picoammeter (Keithley 485/4853) and a low voltage capacitance meter (ICE
electronics model l-6). The measurement instruments are connected through an
in-house built printed circuit board (see Appendix B) using switches that have an

open state resistance larger than 1012 Q.

Hahn-a1 lanthan-
Vestal-

 

Figure 4. Fusion of vesicles with a planar BLM. A - scheme of a BLM. B -
initial stage. prefusion state (vesicle attachment). D - osmotic swelling of the
vesicle and fusion. F - final stage: incorporation.

57

 

 

10 pA

10 sec

 

 

 

Figure 5. Current record during reconstitution. The BLM was held at
+40mV. A sharp increase in the current is typical to fusion of a vesicle(s).
Arrows mark vesicle suspension injection. Channel activity can be seen
immediately after the fusion.

 

58

 

  
 

 

Electrodes

pl
,1
,4
a
r4
pl
,4
A
,1
/////////////////////////////H

BLM Setup

 

 

 

 

 

 

 

Figure 6. Measurement system scheme. l. picoammeter, V. electrometer
and C. capacitance meter. S1 and 82 are switches.

All the measurement instruments are connected to a computer (DHK Systems
3868X) via a data acquisition and control system (Keithley 570). The data
acquisition and control system also provides the command voltage. The system
allows the measurements of open Circuit potential, current under voltage Clamp,
voltage under current clamp and capacitance. Clamping of the voltage or current
is done by software. The software (listing in Appendix C) was developed by the
author using a Borland C++ version 3.0 compiler (Borland lntemational lnc.).
Chart recorders were used to plot the capacitance and current traces. Current-

potential (I-V) relationship data was stored on disk.

59
Chemicals

Soy bean total phosphatide extract (SBTPE) was purchased from Avanti Polar
Lipids Inc. All the chemicals were obtained from Sigma Chemical Co. except

diethylsulfide, which was obtained from American Tokyo Kasei Inc.

Experimental Design

The principle of the experimental design is a study of binary mixtures. Mixture
discrimination index (Equation 1) is an expression of the response to the mixture
in relation to the average response to its components. This is feasible when
there is a complete ensemble of the receptors. In a reconstituted system there is
only a probability of reconstituting the appropriate receptors. The probability of
incorporating more than one receptor is a compound probability and for lack of
evidence was assumed to be a product of independent probabilities. Unlike the
recordings of electrical responses from a large number of ORCS, the
reconstituted system is expected to have few receptors. This fact prevents a
quantitative comparison of the responses. The responses therefore were
analyzed qualitatively in relationship to responses to other odorants. The
possible qualitative responses to a pair of odorants are: no response to both (-),
no response to the first but response to the second (-+), response to the first but
not to the second (+-) and response to both (H). The first case (-) indicates the
absence of receptors to both odorants. The second case (-+) indicates that the
receptor for the first odorant has not been incorporated but a receptor activated
by the second odorant is present. The fourth case (++) indicates that the two
odorants induced a response. Two situations can cause this kind of response:
(a) the second odorant activated the same receptor to a larger degree and (b)
there are two different receptors. It is impossible to differentiate between the two

because the second odorant might be a full agonist while the first one is only a

60
partial agonist. In the third case (+-) it is impossible to determine if the second
odorant activated the receptor activated by the first odorant. The only conclusion
that can be derived is that the second odorant did not activate a receptor
independent of the one activated by the first odorant. However, a response of
this kind might provide an answer about the possibility of full and partial
agonists. If the first odorant presented is the one suspected to be the partial
agonist then the addition of the full agonist should result in additional
conductance Change (response type ++). Furthermore, the presentation of the

partial agonist before the full agonist should never result in a response of the

M39 (-+)-

 

Diethylsulfide Thiophene
HaC—CHZ —-S—CH2 —CH3 | l

Diethanolsulfide
Ho-H2 C—CH2 --s-CH2 —CH2 —0H

 

 

 

Figure 7. Chemical structure of the odorants

As discussed in the introduction, three odorants were used. It is known that
diethylsulfide (DES) activates a receptor that increases K+ conductance.
Thiophene (THP) induces hyperpolarization. Thiophene, which shares the
functional group -CH2-S—CH2- with DES, differs from it in its physicochemical
properties (thiophene is a rigid ring molecule but smells similar to DES).
Diethanolsulfide (DOS) has the same -CH2-S-CH2- group but has a distinctly
different odor. DOS is essentially DES with -OH groups at the ends. With

61

respect to mixtures and osmophores, a mixture of DES and THP was not
expected to present a mixture of osmophores. On the other hand, a mixture of
DOS and DES was expected to contain at least two different osmophores.

Based on the structure of the odorants the following observations were

 

 

 

 

 

 

 

 

 

 

 

expected' :
Table 5. Expected responses to different pairs of odorants.
Odorant pair Expected responses
DESQDOS -4. ’ 4.- ’ ++
THPCDDOS -+ ’ 4,.-. ++
DOSQDES 4.-
DOS cRTHP +-
THPCPDES +-
DES IEOTHP ' +-
am

All the experiments started with a stage of reconstitution of vesicles into a
BLM. The vesicles are fragments of plasma membrane of olfactory mucosa
neuron cilia. The membrane isolation protocol used in this study was shown to
produce ciliary membrane fragments with some cytoskeleton contamination. In

the frogs olfactory mucosa the only cells that have cilia are the receptor neurons.

 

' (--) is expected from all the protocols.

62

A successful reconstitution was determined by a step increase of the current
while the BLM is held at +30 to +50 mV (with respect to trans side). If no
accompanying capacitance change was measured, the current increase
indicated a fusion of a vesicle(s) with different conductance properties than the
BLM. An osmotic gradient was maintained across the BLM to enhance fusion
rate. Reconstitution was concluded by eliminating the gradient by addition of
glycerol to the trans side.

The bathing solution used in all the experiments is: 50 mM Na acetate, 50 mM
K acetate, 5 mM HEPES at pH 7.4. The acetate- anion of was chosen as an
impermeable anion to eliminate the large chloride conductance. Experiments
were done at ambient temperature (around 27°C). 10 mM Ca acetate was added
as required for the fusion of the vesicles. 0.35 M glycerol was added to the Cis
side to create an osmotic gradient across the BLM and across the vesicle
membrane (the vesicles were loaded with 0.4M glycerol). After reconstitution,
glycerol was added to the trans side.

In general, two odorants were sequentially added. When the second odorant
did not induce a conductance change, a third odorant was added. The reason
for the addition of the third odorant was the possibility of lack of receptors for the
second odorant in the particular sample.

Most experiments used a similar sequence of steps. These steps were:

. Obtaining the olfactory cilia membrane fragments.

. Formation of a BLM.

. Addition of osmoctant (glycerol) to the Cis side.

. Injection of the vesicle suspension to the Cis side while monitoring the
current.

. Stopping the reconstitution by addition of osmoctant to the trans side.

63

. Addition of an agent (odorant , inhibitors etc.) and measurement of
conductance and capacitance.
The last step was repeated as necessary.

There are 6 ordered combinations of two out of three odorants. In order to
compare between the responses of the system to a single odorant and a mixture,
an ordered pair has to be presented. All six combinations were tested (Table 6).
The third odorant was presented when no response was observed to the

addition of one of the first two.

 

 

 

 

 

 

 

Table 6. Odorant introduction protocols.
Protocol type A B C
1 DES DOS THP
2 DES THP DOS
3 DOS DES THP
4 DOS THP DES
5 THP DES DOS
6 THP DOS DES

 

 

 

 

 

 

90_nt£>.|§

The controls for a reconstitution experiment are intended to show that the
measured increased conductance was due to the incorporation of vesicles
containing relevant biological material. Thus the control has to Show that without
the fused vesicles the compounds added to the bathing solution do not generate
the same responses. A second control is for the delivery vehicle of the

compounds. Since the odorants are hydrophobic they were delivered in

54

dimethylsulfoxide (DMSO). The second control has to show that DMSO does not

generate the same observed responses as DMSO + odorant.

The following are experiments that serve as the described controls:

(a) Measurement of conductance and capacitance changes in response to all
the added compounds but without reconstitution.

(b) The experiments in which there was no response to odorants serve as an
additional control to show that not all vesicles contain the specific receptors
needed.

(0) Measurement of conductance and capacitance changes in response to
DMSO with and without reconstitution.

Figure 8 shows the effects of DES and THP on a BLM. The addition of more
than 0.1 (M of each of the odorants did not result in a conductance change
above the noise level. An obvious fusion event was observed before the
odorants were added. The integrity and functionality of the reconstituted BLM
were supported by the blocking effect of 4-aminopyridine (300 nM, not shown in
the figure). A similar experiment is shown in Figure 9. In this experiment as in
the one shown in Figure 8, a Clear fusion event was observed before the
odorants were added and 4-AP reduced‘the conductance (not shown). Figure 10
shows an experiment that tested the effect of DMSO. DMSO was added to about
20 times the maximal concentration used in a normal experiment with no
apparent effect. The conductance Change was small and did not show a trend. In
general, no response to odorant was detected with unmodified BLMS even at
concentrations more than ten times higher than the saturating concentrations.
The majority of the reconstitution experiments did not show a response to the
odorants, indicating the requirement for the appropriate receptors and the lack of

direct effect of the odorants on a BLM after reconstitution.

65

 

 

 

 

30.000 ~
:3. 20.000~
G)
U
G 10.000 —
C0
1'3
0.000 ~ 0/0
:5 X*\D'/O
PU // \\\\O
r: 4391/ /
0 —10.000 — ‘
C)
‘53 —2o.ooo-
Z
—30.000 ~
—10 —9 —8 —7
Lo odorant
o DES g[ ](1Og M)
- THP
Figure 8. Effects of DES and THP on BLM. No significant change of

conductance was observed.

 

 

 

 

30.000 ~
3%. 20.000»
<0
E 10000 "No/0
(D ° 0/ \\ /,,/+--———-
-i—) ..
S 0.000i /
E
O —10.OOOL
O
E —20.000~
Z
—30.000 ~
—10 —9 —8 —7 —6
Lo odorant
o THP g[ ](log M)
0 DOS
Figure 9. Effects of THP and DOS on BLM. No significant change of

conductance was observed.

67

 

 

 

 

 

30
EB 20 ~
3'
10 _ O
a \.
E 0
E
CD
’ —iO -
E
C10 —20 ~
-—:’>0
[DMSO] (log M)

Figure 10. Effect of DMSO on a BLM. No significant change of
conductance was onserved.

68

Inhibitors are commonly used to identify and isolate the contribution of
different currents to the observed total current. Inhibitors can be used in
reconstitution experiments where Single channel activity is studied. For example,
4-aminopyridine (4-AP) was shown to block the DES-gated channel (139). In a
large scale reconstitution, the reduction of current by inhibitors is of a lesser
value. Inhibitors like 4-AP act on quite a few types of channels. Since the ion
Channel composition of the vesicles is not defined, 4-AP or other inhibitors might
affect the Channels activated by an odorant as well as other Channels. A
reduction in conductance therefore cannot be attributed unequivocally to
inhibition of the odorant-activated Channel(s). The reduction of the conductance
might be due to inhibition of the odorant activated Channel(s) but can also be
due to inhibition of channels contributing to the baseline conductance (the
conductance after fusion but before the addition of odorants). Inhibitor effects
are informative regarding the condition of the BLM. In particular, ionic inhibitors
like CoSO4 serve to test for a leakage. Since such inhibitors are electrolytes and
are added in concentrations above 10mM, leakage is manifested in increased

conductance rather than inhibition.

EOG recording

Animals were anaesthetized and pithed as described before. The olfactory
epithelium of one nasal cavity was exposed by removing the dorsal skin and
bone. The dorsal part of the epithelium was carefully removed. The ventral part
of the epithelium was then rinsed with a solution containing the electrolyte
composition of the mucus (52.7 mM NaCl, 10.6 mM KCI, 5.35 mM CaCl2 and 5
mM HEPES pH 7.0). Care was taken to prevent blood from contacting the
exposed surface of the epithelium. The animal was then secured to a stable

platform and a tube was inserted through the internal naris to the lateral side of

69

the cavity. This tube was used to evacuate the odorant and other solutions from
the exposed nasal cavity. A reference electrode (Ag\AgCl) was inserted under
the skin in the oral cavity. A recording electrode (glass pipette filled with 0.5 M
KCl connected via a Ag\AgCl, z 10 K9 ) was positioned gently on the surface of
the epithelium in the area of the olfactory eminence. Upon contact, a potential of
around - 40 mV was usually detected. A scheme of the setup is depicted in

Figure 11.

Odorant delivery pipette

Recording electrode
Nasal Cavity

  
  

Olfactory eminence

Figure 11. An illustration of the setup used to record EOG.

The system was left to stabilize for about 15 minutes and then the odorant
solutions were introduced (odorants in DMSO). EOG has two phases: a phasic
potential followed by a tonic one. Since the odorants were introduced in the
liquid phase the stimulation resembled a constant stimulation in that the EOG
did not go back to control but rather stayed at the last tonic potential until the
epithelium was rinsed.

Dose-response relationships were measured by recording the potential

change due to increasing concentrations of odorants. The same volume of

70

odorant solution was introduced each time. After the peak of the phasic
response the excess solution was evacUated.

MDI recordings were done using concentrations below E050 for both
odorants. The responses to each of the odorants in the chosen concentrations
were compared to the response to a mixture of the two. The mixture contained

the same odorants in the concentrations that were tested individually.

Analysis

Current-voltage data analysis was done using a program developed by the
author. The program fits a straight line to the data using either linear regression
or robust statistic (M-estimate). Parts of the data set can be analyzed separately.
The program was developed using a Borland C++ version 3.0 compiler (Borland
lntemational Inc.) and used adaptation of procedures described by Press et al.
(97). The slope of the l-V data was taken as the conductance. Figures 12 and 13
are examples of the recorded data before analysis. The conductance of the BLM
after a reconstitution event was regarded as the response to zero concentration
of odorant and thus subtracted from the conductance values measured after
each of the treatments.

Dose response curves were analyzed using the four parameter logistic model
(27). This model has been used extensively to describe data from bioassays,
radioreceptor assays and radioimmunassays (27, 37, 141, 84, 52). Equation 2

describes the four parameters logistic model.

71

 

300 -
’0
200 ~ /
/
100 ~ / .
A / ‘
<5. . ,
v 0 _... -....._—— ~— , “r” ‘—
E _40 —20 xv 20 4O
. / Vm (mV)
_1 00 " " .
—2oo - /
-300
o BLM I 23nM DES
O Reconstitution A 123nM DES
v lnM DES A 300nM 4—AP
V BnM DES <> 13mM CoSO4
o 13nM DES
Figure 12. Voltage-current relationships measured in an experiment to

determine a dose-response relationship. The conductance of the modified BLM
was calculated as the slope of each potential-current relationship.

72

 

 

 

 

30 r
Cl
20 . /
/.
/W/
20 40
m (mV)
0 BLM I 300nM 4—AP
O Reconstitution A 13mM CoSO4
v IOOnM THP
v IOOnM DES
D luM DOS
Figure 13. Voltage-current relationships measured in an experiment to

determine responses to the three odorants. The conductance of the modified
BLM was calculated as the slope of each potential-current relationship.

73

Equation 2.

Where : Y is the response; X is the dose ( in logarithmic scale); a is
the response when X = 0; dis the response when X = oo; 0
corresponds to ED50 i.e. the dose at half maximal response and b

is the slope factor.

The parameters were fitted to the data points using a non-linear least square
minimization algorithm (Levenberg-Marquardt). An interactive computer program
developed by the author allowed constraints to be imposed on the fitting. Any of
the parameters (up to three out of the four) could be held constant. The program
also provided the 12 value, the probability of 1-a (statistical significance), ED5%
and EDg5%.

To pool data from different experiments, a normalization procedure was used.
The data from different experiments vary in the magnitude of the conductance
change (depending for example on how many channels were present) and
possibly in the concentrations of odorants used. The data were normalized by

the values of the parameters a and d (baseline and maximal conductances). The
conductance values were transformed by Y' = gif— . Thus, a is taken as 0.0

and d as 1.0. After transformation the data could be pooled together provided
that the parameters of the individual curves were similar. In the case of curves
with divergent parameters, the fitted curve of the pooled data has a low p value.
When a curve was fitted to the pooled data, the parameters a and d were held at
0.0 and 1.0 respectively. Imposing these constraints is analogous to the
procedure described by De Lean et al. (27) and Guardabasso et al. (52). As

these authors explain, the constrained four parameters logistic model contains

74

more information than the unconstrained analysis especially when the data are

biased toward a particular region of the curve as well as for curves with sparse

data. Following Press at al. (97) the individual variances (020 are estimated as
[y. -y(Jr,-)]2

N
0’2 = 2T' Thus, equal contributions of variance are assumed for each
i=1

of the data points. The variance is adjusted after each call to the minimizing
routine. ‘

In analyzing the responses of a reconstituted system to saturating
concentrations of odorants the criteria for a response were: (a) the conductance
increase was larger than 3% of the modified BLM (reconstituted system)
conductance and (b) the conductance increase was at least 35pS. Three percent
is about twice the noise level of a high resistance modified BLM and is at the
same time below the contribution of a single 35pS channel in a usual
reconstitution (about 0.5 nS). The second criterion, conductance increase of at
least 35pS, is based on the smallest conductance of an identified ion channel
that responds to odorants directly.

EOG potentials were normalized by a similar method. The response to DMSO
was subtracted from each of the responses and then the net maximal response
was taken as 1.0. Dose response curves were processed similarly to the dose-

response in the reconstituted system.

RESULTS

The main thrust of the experiments performed in this study was to record the
responses of reconstituted olfactory cilia membrane (OCM) to two odorants
presented one after the other. Since the odorants had to be introduced at
saturating concentrations, dose—response relationships were established. To test
the relationship between DES and DOS in the intact system, liquid phase EOGs
were measured and MDls were calculated. This section presents data from the
dose-response curves, examples of the responses to pairs of odorants, EOG
and MDI measurements and the frequencies of the responses to each of the

odorants and to pairs of odorants.

Dose-response curves

Dose-response curves of the three odorants were measured. The data are
presented as normalized conductance. The normalization is described in detail
in the methods section. Briefly, a curve was fitted to the data from each
experiment. The curve was constructed using the Four-Parameter Logistic
model. Using the model effective doses (ED) were calculated. Two of the four
parameters describe the minimal and maximal responses. The difference
between these two parameters is the net magnitude of the maximal response to
the stimulus. The estimated parameters were used to normalize the data. First
the minimal response was subtracted from the data. The results of the

subtraction are the net changes due to the stimulations. Then, the net changes

75

76

were divided by the difference between the maximal and minimal responses. The
division transforms the data to the range 0.0 to 1.0. The transformation affects
only the y-axis and thus allows the combination and comparison of data from
different experiments. The model was also used to calculate the ED5, E050 and
E095. ED5 was taken as an estimate of threshold concentration. ED95 was
taken as an estimate of the saturating concentration.

Figures 14, 15 and 16 show the dose-response relationships for the three
odorants. The data points from different experiments are represented by different
symbols. The following was calculated from the dose-response curve for DES:
E05 = 1.05 nM, ED50 = 2.63 nM and E095 = 7.59 nM (x2 = 0.509; p = 1.00
where p is the significance level). The following was calculated from the dose-
response curve for DOS: ED5 = 83.2 nM, E050 = 126 nM and ED95 = 195 nM (
x2 = 0.130; p = 1.00). The following was calculated from the dose-response
curve for THP: ED5 = 0.28 nM, ED50 = 0.71 nM and E095 = 2.19 nM (x2 =
0.01; p = 1.00).

The values calculated for DES are in agreement with the observed threshold
at the single channel level (139). Figure 17 illustrates the differences between
the three odorants. Note that although there are considerable differences in the
location of the curves along the concentration axis, the dynamic range of each of
the odorants is about 1 logarithm unit. This observation agrees with the data
from isolated ORCs (39, 38). The comparison of threshold concentrations
(ED5%) of THP and DES is in agreement with the observation of thresholds in
humans. In Humans the threshold for detection of THP is three times lower than
that for detection of DES (36). Table 7 lists the calculated effective dose for 5%,

50% and 95% of maximal response.

Normalized Conductance

77

 

1.2~ ‘ -

 

 

 

 

—-lO --9 -8 —7

Figure 14. Dose-response curve of reconstituted OCM to DES.
Pooled data from three experiments. x2 = 0.509; p = 1.00; E05% =
1.05 nM; 505%, = 2.63 nM; 509596 = 7.59 nM.

Normalized conductance

78

 

0

 

 

 

—lO —9 e8 ~7 —6 —5

Log[DOS](LOg M)

Figure 15. Dose-response curve of reconstituted OCM to DOS.
Pooled data from two experiments. 2:2 = 0.130; p = 1.00; 505% = 83.2
"M; 505096 = 126 nM; 509596 = 195 nM.

Normalized conductance

79

 

 

 

 

 

—lO —9 -8

Log [THPLIOg M)

Figure 16. Dose-response curve of reconstituted OCM to THP. x2
= 0.01; p = 1.00; 505% = 0.28 nM; 505096 = 0.71 nM; 509596 = 2.19
nM.

(D
o 1.0
I:
.53
C.) 0.8 ”
:3
"S
o 0.6 —
C)
"U
,1, 0.4—
B
E 0.2-
SH
2
0.0
Figure17.

odorants.

80

 

 

 

 

 

 

THP DES DOS
.
—lO —9 —8 —7 -6
Log[odorant](log M)

Comparison of the Dose-response curves of the three

81

Table 7. Effective dose (ED) values for the three odorants.

505% (nM) EDSO% (nM) 5°95% (nM)
THP 0.28 0.71 2.16
DES 1.05 2.63 7.59
003 83.2 126 195

In some experiments the full range of concentrations was not measured (the
membrane broke before the experiment was complete). The threshold was
estimated from these experiments as the average of the first concentration to
induce increased conductance and the concentration preceding it. The estimated
value for the threshold for DES is 1.34:0.34nM (SE 1' s. E. n = 10). Similar
estimation of the saturating concentration of DES is 8.38:0.51nM (i i s. E. n =
4). These values are similar to the values produced by the logistic model
although they appear to be slightly higher.

Some of the observed dose-response relationships are not monophasic. An
example is a dose-response curve for THP that shows two regions of lesser
slope (Figure 18). The data from this experiment was analyzed by partitioning
the data to two subsets. The lower concentration was fit to a curve similar to that
shown in Figure 16 while the rest of the data points were analyzed
independently. The sum of both curves fits the data points. This type of dose-
response is interpreted as two classes of binding sites (84). In the case of
concentration-conductance relationship this interpretation is not exclusive

because of other possible mechanisms such as change in conductance state.

82

 

 

 

 

Normalized Conductance

 

 

 

 

— 1 0 — 9 — 8 — 7
Log[THP](log M)
Figure 18. Dose-response of reconstituted OCM to THP. The data was fit to

two subsets of the data. A - Lower few points were fit using the same parameters
as in fig 16 ; B - Upper subset was fit independently: ED50 = 3.55 nM, ED5 =
2.88 nM and E095 = 4.37 nM; C - the sum of A and B.

83
Responses to pairs of odorants

As mentioned above, the main thrust of the experiments performed in this
study is the comparison of the responses to pairs of odorants introduced one
after the other. The three odorants were introduced at saturating concentrations.
The results of the experiments were analyzed qualitatively: did the system
respond to the presented odorant? Or did it not? In part of the experiments the
values of the conductance changes were small enough to allow suggestions
about the conductance of the ionic channels that underlay the response to the
odorant. The figures shown below are examples of experiments that resulted in
the three types of responses.

Since the odorants were presented at saturating concentrations, the
responses are indicative of the availability of a receptor. The responses of
reconstituted OCM to the introduction of a sequence of two odorants can be
divided into three types. The first type is a response to both odorants. The other
two types of responses are the results of those experiments in which the
reconstituted OCM responded to only one of the odorants. In one case the
system responded to the first odorant but not the second and in the other case
the reverse occurred. I

Figure 19 shows a response to DES and an additional response to DOS.
CoSO4 and 4-AP blocked some of the conductance. The odorants and the
blockers induced large conductance changes although the membrane
conductance was relatively low after the fusion (76.8 pS). A clearer way to
present the effects of each of the treatments (odorants and inhibitors) is to
display the changes of the membrane conductance between each of the
subsequent treatments. The net effect of each of the compounds presented to
the reconstituted system is thus taken as the change of membrane conductance

from its last value. The net effects of the treatments in the experiment shown in

84
Figure 19, are shown in Figure 20. A response of the same type is shown in
Figure 21. The reconstituted OCM responded both to THP and to a subsequent
introduction of DOS. Similar to Figure 20, Figure 22 shows the net effects of the
odorants and inhibitors that were presented to the reconstituted OCM. The
conductance changes in this experiment were of a magnitude of a few ion
channels. THP induced a membrane conductance increase of 130 pS and DOS
of 80 pS.

The second type of response was a response only to the first odorant that
was presented. Figure 23 shows a response only to DOS. A saturating
concentration of DOS was presented inducing a conductance increase of 80 pS.
The addition of a saturating concentration of DES did not change the membrane
conductance significantly. Figure 24 shows the net effect of the odorants and a
stepwise conductance block by 4-AP. A similar response was observed when
THP was introduced after DOS. Figures 25 and 27 are examples of experiments
where DOS was presented before THP inducing a conductance change but THP
that was added subsequently had no effect. In the experiment depicted in
Figures 25 and 26, DOS induced a membrane conductance increase of 332 p8.
In the experiment depicted in Figures 27 and 28, DOS induced a membrane
conductance increase of 319 pS.

A third type of response is a response to the second odorant but not the first
one (Figure 27 and 29). In the experiments exemplified by Figures 27 and 29,
DES (Figure 27) or THP (Figure 29) were introduced before DOS. The
reconstituted OCM did not respond to the first odorant but responded to the
addition of DOS. In the experiment depicted in Figures 27 and 28, DOS induced
a membrane conductance increase of 319 pS. In the experiment depicted in

Figures 29 and 30, DOS induced a membrane conductance increase of 80 pS.

85

 

 

 

 

 

 

 

 

 

 

 

Net Conductance
(HS)

 

 

 

 

 

 

 

 

 

 

 

— + 135nM DES
+ 25,uM Dos
— + BOOnM 4—AP
— + 10mM Coso4

UOCDZI>
|

Figure 19. Response to both DES and DOS. The membrane
conductance after fusion was 76.8 pS.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

4
”m"
5
CD
0
C 2
(t5
.4.)
C)
:5
"U
a: O
O
U
4.)
(D
Z
-——2
A B C D
A — + 135nM DES
B — + 25,uM DOS
C — + BOOnM 4—AP
D — + 10mM CoSO4
Figure 20. Membrane conductance values from Figure 19 plotted as the

difference between successive treatments (the net effects of each of the
treatments).

87

250 1 .

 

 

ZOO

 

 

150

 

 

 

100

50

 

 

 

 

 

 

 

O

 

—SO

 

Conductance Change(ps)

 

—lOO

 

 

 

 

 

 

 

 

L.

 

A B C D E
100nM THP

20nM ATP+GTP

luM DOS

BOODM 4—AP
13mM COSO4

WUQCD3>
|

Figure 21. Responses to THP and DOS. The membrane conductance after
fusion was 1.57nS

88

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

200 T f T T T
a; 100
8;
(1)
Q0
:3 0
CU
Ll
U
m —100
C)
c:
(D
*5 —200
:3
"U
c:
o —300
U
~4oo 1 1 . . .
A B C D E
A — IOOnM THP
B — 20nM ATP+GTP
C — 1,uM DOS
D - 300nM 4—AP
E — 13mM CoSO4
Figure 22. Membrane conductance values from Figure 21 plotted as the difference

between successive treatments (the net effects of each of the treatments). A - THP
induced a conductance increase of 130 ps; 8 - ATP and GTP did not have an effect; C -
DOS induced additional conductance change of 80 pS; D - 4-aminopyridin mduced the
conductance by 340 pS; E - CoSO4 did not have an effect.

89

 

100 . 4

 

 

80

 

60

 

4O

 

 

20

 

 

 

 

 

 

 

 

 

l |
em
00

 

Conductance Change (p8)

 

 

 

 

 

 

|
on
o

i

LSMM DOS
lOOnM DES
150nM 4—AP
ISOnM 4—AP

c0w>
l
++++

Figure 23. Response to DOS and no response to a subsequent
addition of DES. The membrane conductance after fusion was 438 pS.

90

 

100 . ' . - .
80

 

 

 

60
4O

 

 

N
O

 

 

 

 

 

 

 

 

Conductance Change (ps)
0

 

 

 

l I | I
oo 03 4s l\)
O O O o

 

—1OO

 

 

LSDM DOS
lOOnM DES
150nM 4—AP
150nM 4—AP

0000:»
l
++++

Figure 24. Membrane conductance values from Figure 23 plotted
as the difference between successive treatments (the net effects of
each of the treatments). A - DOS induced 82 ps; 8 - DES changes the
conductance by -6 ps; 0 - 150 nM +aminopyridine blocked 39 pS; D -
increasing the concentration of 4-aminopyridine blocked additional 81
pS to a total of 130 pS.

91

350 1 ‘s . .

 

 

BOO

 

 

250

 

 

200

 

 

150

 

100
50

 

 

 

 

 

 

 

Net Conductance
(p8)

 

 

 

 

_50 t , t L J

100nM DES
l,uM DOS
lOOnM THP
BOOnM 4—AP

onw>
l
++++

Figure 25. Response only to DOS. DES was introduwd before
DOS and THP was introduced after DOS, both with no effect. The
membrane conductance after fusion was 126 pS.

92

4430

 

 

3CH)

 

 

ZCN)

 

1()O

 

 

 

(knuhufiance Change(p$

 

 

 

 

 

 

—1OO

IOOnM DES
luM DOS
100nM THP
300nM 4—AP

UGOJID
ll
++++

Figure 28. Membrane conductance values from Figure 25 plotted
as the difference between successive treatments (the net effects of
each of the treatments). A - DES induced a -20 pS change; 8 - DOS
induced a 332 p8 change; C - THP induced -18 pS change; D - 4—AP
induced a -69 pS change. '

93

350 . . . .

 

300

 

250

 

 

 

200

 

150

 

100
50

 

 

 

 

 

 

 

 

Net Conductance
(p8)

 

_SO 1 t i m

 

 

lllnM DES
400uM DOS
100nM THP
BOOnM 4—AP

o0m>
l
++++

Figure 27. Response to DOS without a response to a subsequent
addition of THP. The membrane conductance after fusion was 147 pS.

94

 

400 . , 1 .

 

 

300

 

200

 

IOO

 

 

 

Conductance Change (p8)

 

 

 

 

 

 

--lOO

IllnM DES
400uM DOS
100nM THP
BOOnM 4—AP

U0w>
|
++++

Figure 28. Conductance values from Figure 27 plotted as the
difference between successive treatments (net effect of each
treatment). A - DES induced a -20 pS change; B - DOS induced a 319
pS change; C - THP did not have an effect; D - 4-aminopyridine
reduced the conductance by 74 pS.

95

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

50
0
”a?
8;
<1)
0
g 50
_,_, _
O
:3
"U
8
C.) — 1 OO
4.)
0)
z
—150 e .
A B C D E F
A — +100nM THP D - + luM DOS
B — + 20nM ATP+GTP E - + 300nM 4-AP
C — + 100nM DES F — + 13mM CoSO4
Figure 29. Net membrane conductances when the only response was

to DOS. THP and DES which were introduced before DOS had no effect.
The membrane conductance after fusion was 553pS.

 

100 w T

 

SO

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Conductance Change (p8)

 

 

 

—50
—100
—150 .
A B C D E F
A — +100nM THP D - + luM DOS
B — + 20ml ATP+GTP E - + 300nM 4-AP
c - + 100nM DES F - + 13mM C030,,
Figure 30. Membrane conductance values from Figure 29 plotted as the

difference between successive treatments (depicting the net effect of each of the
treatments). A - THP induced a -16 pS change (note: within the noise level); 8 -
ATP and GTP induced a -31 pS change; C - DES induced a -2 pS change (note:
within the noise level); D - DOS induced a 80 pS change; E - 4-AP induced a -55
pS change; F - CoSO4 induced a -101 pS change.

97

The examples provided above illustrate the responses of reconstituted OCM
to the presentation of pairs of odorants. The reconstituted OCM system
responded to one of the odorants or to both. After the system responded to
DOS, no additional responses were observed when either DES or THP was
introduced (n = 39). Additional responses were observed when DOS was
introduced following DES or THP.

In some of the experiments the values of the membrane conductance change
as well as the values of the membrane conductance blocked by inhibitors allows
speculation about the underlying ion channels. The membrane conductance
changes observed in all the experiments could be fit (either as single or a
multiple) to the conductance of ion channels identified in ORCs (table 2). The
results of the experiments where the change of membrane conductance was
relatively small indicate that DOS induced conductance increases of 80 pS or 35
p8 (and the multiples). The membrane conductance increases induced by DES
were around 60 p8 (and the multiples). No attempt was made to characterize the
ion channels. The values of the membrane conductance change did not differ
when K+ was the only cation compared to experiments where both K+ and Na+
were present (Ca2+ was present in all the experiments).

Three inhibitors were used: ruthenium red (RR), 0080.; and 4-aminopyridine
(4-AP). Since the OCM vesicles contain a variety of ion channels the effects of
the inhibitors can not be used to identify the ion channels underlying the
response to odorants. As mentioned in the methods section, the inhibitor effects
can serve as a test of the integrity of the BLM. In particular, the effect of C0804
serves as a gauge for leakage. Since CoSO4 is an electrolyte and was added in
concentrations above 10 mM, it is expected to increase the conductance in a
case of a leak. CoSO4 normally blocked about 80pS or multiples of 80pS. One
mM 00804 was reported to reduce a Ca2+-activated K+ current (83). The

98

presence of a Ca2+-activated K“ channel with a conductance of 80pS was
reported (80). Furthermore, Co forms a partially covalent bond with sulfur and
thus chelates sulfur containing compounds (77). Since all three odorants are
sulfides, CoSO4 might reduce the free odorant concentration substantially.
Experiments in which application of 00804 did increase the conductance were
discarded. Effect of 4-AP was observed in all the experiments where the
reconstituted OCM responded to DES, THP or DOS. CoSO4 did not exhibit such
a uniform effect: in some experiments it reduced the membrane conductance but
in others it had no effect, even when an effect of an odorant was observed.
Cyclic-AMP and IP3 are thought to mediate odorant transduction. GTP (20
uM) was added as it is required for the activation of AC and PLC by a receptor.
ATP (20 uM) was added as a substrate for AC. Based on the literature, it is
possible that in some of the experiments the lack of response was due to the
absence of ATP or GTP. ATP and GTP were added to the reconstituted system
after odorants, particularly in those experiments where no response was
observed after the addition of the odorants. No increased conductance was
observed after the addition of ATP and GTP. In some experiments the
membrane conductance was reduced (usually by about 20 to 30 pS) after the
addition of ATP and GTP. This observation conflicts with a previous report
(135). Cyclic-AMP mediated transduction was not reported to be active in the

range of odorant concentrations used in this study.

99
Frequencies of responses.

A qualitative analysis of the responses to odorants is concerned only with a
'yes or no' answer to the question: did the system respond to the odorant? The
magnitude as well as other features of the response are not relevant to the
analysis. As described in the methods section, the criteria used to classify the
responses were: (a) the response had to be at least 35 pS and (b) the response
had to be at least 3% of the conductance of the system before odorants were
added. In order to eliminate the dependence of the response on odorant
concentration, the odorants were presented at concentrations greater than their
saturating concentrations.

Except for responses of marginal magnitude, the response of reconstituted
OCM to saturating concentrations of odorants depends on the availability of the
receptors for these odorants. Each reconstituted BLM is a product of sampling
from a large pool of vesicles containing various receptors. Since the deciliation
process is not biased toward any particular ORC or receptor, it is assumed that
the resulting OCM vesicles contain a random arrangement of receptors and ion
channels. It follows, that each reconstitution experiment is a random sample of
receptors. Table 8 lists the numbers of the responses to each of the odorants.
The responses are also divided according to the place of the odorant in the
sequence of presentation. Reconstituted OCM did not respond to DES or THP
when these odorants were presented after another odorant. Only DOS induced

responses both as the first and the second presented odorant.

100

Table 8. Frequencies of all the responses to odorants. The
responses are not divided in relation to other odorants that were tested. The
responses are further subdivided into those that were produced when the
odorant was tested first. Responses to DOS are further subdivided into those
that were produced when DOS was presented as the second odorant. DES and
THP did not elicit a response when presented second. The last two rows total
responses for the three odorants: a subtotal of the responses to the odorants
when presented as the first odorant and the grand total of all the responses. (N
tested) are the total number of experiments where the odorant was tested; (‘7)
are the number of unclear responses to the odorant; (N clear y/n) are the
number of experiments where a clear response or no response were observed.
(Yes) are the numbers of responses to the odorant. (No) are the number of no
response to the odorant; (Frequency) the proportion of positive responses from
these experiments where clear responses were observed.

 

 

 

 

 

 

 

 

 

 

 

Odorant N ? N clear Yes No Frequency
tested yln
DES 89 1 1 78 16 62 20.5%
DES as 1st 54 16 38 29.6%
THP 66 6 60 9 53 15.0%
THP as 1st 33 9 24 27.3%
DOS 64 7 57 20 37 35.1%
DOS as 1st 21 11 10 52.4%
DOS as 36 9 27 25.0%
2nd
Total 108 36 72 33.3%
response
to 1st
Total 219 24 195 45 152 23.1%

 

 

 

 

 

 

 

 

Table 9 lists the responses to the six pairs of odorants (three odorants in both
orders). The table lists the number of each of the four observed outcomes of
testing each of the odorant pairs. The number of experiments in table 8 is larger
than the number in table 9 because not all experiments tested pairs of odorants.

Thus the frequency of a response to a particular odorant from table 9 is based

101

only on those experiments where two odorants were tested.

Table 9.

Responses to pairs of odorants. (- -) No response to

either. (-+) No response to the first but response to the second odorant. (+-)
response to the first odorant without additional response to the second odorant.
(++) Response to the first odorant and additional response to the second one. (w

) stands for 'followed by'.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. - - + + - + + Total

DESwDOS 20 4 2 1 27
THP¢DOS 14 3 0 1 18
DOSwDES 14 0 5 0 19
DOS wTHP 14 0 6 0 20
THP©DES 13 0 2 0 15
DES©THP 17 0 2 0 19

Total 92 7 17 2 1 18

% of 78% 5.9% 11.4% 1.7%

experiments

 

In some experiments three odorants were introduced. Those were mostly

experiments in which the reconstituted OCM did not respond to the second

odorant. In all of these experiments the reconstituted OCM did not respond to

the third odorant as well. Actually the only odorant that induced a conductance

change as the third odorant was DOS. DES and THP never induced a

 

102

conductance change when introduced as the third odorant. ATP and GTP (20
pM each) were added at different stages of the experiment. No response to ATP
or GTP was observed. The addition of ATP and GTP did not have a consistent
influence on the response to the odorant that followed it. Responses to odorants
were observed without the presence of ATP and GTP. ATP and GTP were
present in 17 out of the 26 experiments (Table 9) where responses to odorants
were observed (65%). ATP and GTP were also present in 63 out of the 92
experiments were no response to odorants was observed (68%). Because of the

lack of effect, the role of ATP and GTP was not taken as a factor in the analysis.

1 03
EOG and MDI measurement

In order to test the for competitive binding among DOS and DES in vivo,
liquid-phase EOGs were recorded and the EOG recordings were also used to
calculate mixture discrimination indices (Equation 1). In order to maximize the
magnitude of the induced transepithelial potential and at the same time avoid
saturation of the potential developed by the ORCs (60), the concentrations of the
odorants were chosen to be just below the E050, which was calculated from
dose-response measurements using the four-parameter logistic model.

An odorant induced transepithelial potential has two phases: a transient
potential and a longer tonic potential. Under continuous vapor-phase stimulation
the epithelium remains in the tonic level. Liquid—phase stimulation produces a
similar response. When the epithelium is rinsed with an electrolyte solution
similar to natural mucus (52.7 mM NaCl, 10.6 mM KCI, 5.35 mM CaClz, 5 mM
HEPES adjusted to pH of 7.25 with KOH) the transepithelial potential returns to
pre-stimulation level (~30 to -40 mV, Ag/AgCll 0.5 M KCl in the recording
electrode and AglAgCl | interstitial fluid as ground). Figure 31 shows a voltage
trace of a dose-response measurement. Liquid-phase introduction of the odorant
produced a potential that resembles a continuous vapor-phase stimulation signal
(5, 1 13).

Some odorants produce a complex simulation. These odorants induce both
depolarization and hyperpolarization of the apical membrane. Figure 31 shows
such a response. The EOG phasic depolarization phase (Figure 31 A) is
preceded by a brief hyperpolarization (Figure 31 B). The initial
hyperpolarization was reported by Laffort et al (64). Similar hyperpolarization
was observed in EOG induced by DES. The hyperpolarization is measurable

only at the nM range.

104

lmV
Al

 

1 min
Bl

DMSO 10'9 10'8 10'7 10‘6 10'5 10'4 10'3

Figure 31. Potential trace of EOG responses to increasing concentrations
of THP; A - Depolarization; B - Hyperpolarization.

The data from the experiment depicted in Figure 31 is presented in Figure 32
in the form of a dose-response curve. The calculated ED5 is about 10 times
larger than that obtained from the reconstituted OCM. The dynamic range
displays a sharp difference from that obtained from in vitro responses. While the
dynamic range of the in vitro measured responses was about one logarithmic
unit, the in vivo measured dynamic range spans about six logarithmic units.

Figure 33 shows an EOG trace when MDI was measured. The odorants were
introduced at concentrations of 1 uM. The mixture of DES and DOS was a
mixture of 1 uM of each of the odorants. 'The responses to 2 pM solutions of the
odorants yielded a slightly higher but not significantly different potential. Since

the effect of DMSO is constant it does not contribute to the calculation.

105

 

Normalized VEOG

 

 

 

 

—9 —8 —7 —6 —5 —4 —3
Log[THP](lOgM)

Figure 32. Dose-response relationship of EOG to THP. E050 is 0.513pM; 12
= 0.038; p = 1.00; E05 = 3.39 nM; E095 = 6.61 mM.

106

5mV l

1 min

luM DOS luM DES DMSO DES + DOS

Figure 33. EOG trace during a MDI of DES and DOS assay.

The averages of 5 MDI measurements are shown in Figure 34. The
transepithelial potentials recorded in the experiments are normalized by dividing

each of the potentials by the average of the potentials induced by each of the
odorants. R,’ =

 

(R +1§ ) ,2 Where R; denotes actual response and Ri'
DES Dos

denotes a normalized response. Average MDI is 1.43 i- 0.037 (mean 1: se.; n =
5). Average RDES is 1.01 :I: 0.057 and average R003 is 0.99 t 0.057 (mean 1:
se.; n = 5). A value of 1.43 (a value between 1.0 and 2.0) is interpreted as a
partial overlap of the activation of the receptors that respond to DES and those
which respond to DOS (60).

107

 

1.60

 

.13

1.40 ~

1.20 ~

 

1.00 ~

 

t——-——1
t—-—t

0.80 ~

0.60 ~

0.40 ~

Normalized VEOG

0.20 ~

 

 

 

 

 

 

 

 

0.00 ‘ : :
DES DOS MIX

 

Figure 34. Average MDI of DES and DOS. The three bars depict
the average normalized EOG responses to: DES, DOS and a mixture of
them. Average MDI is 1.43 i 0.037 (mean :I: se.; n = 5). DES and DOS
were presented at concentrations of 1 nM. The responses are
normalized by expressing Ri /{(RDES + R003) l2}. Average RDES is
1.01 t 0.057 and average R003 is 0.99 :t 0.057 (mean 1: se.; it = 5).

DISCUSSION

This study is based on the multiple profile-multiple receptor site model (96).
The model predicts that odorants might contain in their structure more then one
elementary informational moiety (11). Three odorants were utilized:
diethylsulfide, thiophene and diethanolsulfide. Examination of the structure of
the odorants reveals a common structural part but also shows that DOS contains
an additional potential functional group. Application of the model to the three
odorants yields the prediction that: DOS is capable of activating a receptor that
cannot be activated by DES or THP. It also follows from the model that the
reverse, i.e., activation of a receptor by DES or THP but not DOS, is not likely.

To obtain experimental evidence for these predictions at the responses of
reconstituted OCM to the three odorants were tested. The introduction of each of
the odorant had to be at saturating concentrations. Thus the first stage was to
establish the dose-response relationship of the three odorants in the
reconstituted system. Data from a series of concentrations of an odorant was
analyzed using the four-parameter logistic model, which was used to calculate
effective doses (Table 7). E095 was taken as the saturating concentration. The
odorants were then routinely introduced at concentrations greater than the
ED95. The dynamic ranges of the three odorants are narrow and the three
curves are located separately along the concentration axis (Figure 17). As
expected the affinities of the odorants to the receptors, especially of DES and

THP, are different (36).

108

109

Table 8 lists the frequencies of the responses to each odorant, and
subdivides them according to experiments which the odorant was 1st or 2nd.
The table reveals that response frequencies to each odorant depends on its
place in the order of presentation. Before the statistical analysis of the
frequencies is presented, an explanation of the statistics is due.

The properties of the experimental method satisfy the theoretical
requirements for random sampling. A binary outcome (yes or no) and random
sampling allow the consideration of the frequencies (proportions) of the binary
outcomes as estimates of the probabilities of observing these outcomes.
According to the teory of binomial distribution the proportion of an outcome
equals its frequency and is an estimate of its probability (55, 32).

Since the technique of reconstitution of membrane fragments in a BLM is a
random sample of receptors and since the odorants are presented in saturating
concentrations, the proportion of the experiments in which a response to a
particular odorant was observed reflects (for the most part) the probability of
fusing the vesicles that contain a receptor for that particular odorant.

The abundance of a receptor for a particular odorant can be estimated from
the probability of fusing a vesicle that contains this receptor. A stable Bernoulli
process is defined as a sampling process in which the probabilities are not
changed and there are only two outcomes. Fusion of OCM fragments to a
preformed BLM samples receptors from a large pool. The method requires only
a minute and random quantity of receptors and thus does not change the
probability of finding any specific receptor. This sampling method satisfies the
first condition for a Bernoulli process. The outcomes of introducing a saturating
concentration of an odorant to the reconstituted OCM are either a response or
no response. Thus the second condition of binary outcomes is satisfied.

Satisfying the conditions for a Bernoulli process allows the difference between

110

the proportions to be tested using an approximate confidence interval of
proportions (32, 55). The approximate confidence interval is defined in Equation

3.

Equation. 3.

H05P1=p2 Hlapntpz
Jfil-(t—Dlljz-U-Dz)

 

 

awaken-52)

 

 

51—52131; (p1_p2<51-52+zt—1aJ
2

2 N1 N2 N1 N2
Where: 5,152 is the observed proportions.
N,,N2 is the number of samples in each group .
z is the standardized probability in a normal distribution

a is the significance level

If the confidence interval contains 0.0 then one cannot reject the null
hypothesis (32).

Table 10 shows the values of the confidence intervals for the frequencies
(proportions) of responses to the three odorants. The frequencies were taken
from Table 8. The rightmost column of Table 10 indicates if the frequencies were
statistically different. Totaling the response to odorants without regard to their
position in the presentation sequence might bias the proportion and conflict with
the assumptions made above. Note that both DES and THP did not induce
response when presented second in the sequence. If the odorant is not the first
to be introduced then the response to the odorant might not depend only on the
presence of an appropriate receptor but rather on the availability of this (these)
receptor(s). To avoid this complication the frequencies of response to the
odorants from the experiments where the odorants were introduced first, are
compared. While the frequencies of the responses to DES and THP are not

significantly different, both these frequencies are different from that of the

111

responses to DOS. A comparison was also made between the frequencies of
DES and THP when presented first and the frequency of response to DOS when
it was presented second.

The statistical test shows that the response frequency to DOS differed from
those to DES or THP. This difference occurred both when the total response to
DOS was tested and when DOS was the first odorant to be introduced. The
response frequencies were not found to be different when the responses to DES
or THP were compared to the frequency of responses to DOS when it was the
second odorant to be introduced. Comparison of the total response frequency
should be used cautiously as it is obvious that there are large differences in the
response frequencies to each of the odorants depending on their location in the
sequence of presentation (DES 29.6% vs. 0.0; THP 27.3% vs. 0.0; DOS 52.4%
vs. 25.0%). Note that the difference between the frequency of responses to DOS
as the first odorant and DOS as the second odorant is similar to the frequencies
of responses to DES or THP (52.4% - 25.0% = 27.4% vs. 29.6% or 27.3%). As
will be discussed below, this difference fits other observations that suggest that

DOS activates two (or more) receptor types.

Table 10.

112

Approximate confidence intervals for the proportions from Table
8. Confidence level is 90%(two tail test, a of each side is 5%). DES, THP and
DOS indicate the total number of tests. DES1, THP1 and 0081 indicate the
experiments where these odorants were presented first. 0082 indicates the
experiments where DOS was the second odorant to be presented.

 

Confidence Interval

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Odorants p1 p2 Minimum Maximum Different?
DES-DOS 0.205 0.351 0.018 0.274 Yes
THP-DOS 0.150 0.351 0.073 0.329 Yes
DES-THP 0.205 0.150 -0.052 0.162 No
DES1-DOS1 0.296 0.524 0.022 0.434 Yes
THP1-DOS1 0.273 0.524 0.316 0.470 Yes
DES1-THP1 0.296 0.273 -0.140 0.186 No
DES1-DOSZ 0.296 0.250 A-0.110 0.202 No
TH P1 -DOSz 0.273 0.250 -0.151 0.197 No

 

 

113

To test for competitive binding among the odorants, the odorants were
presented to a reconstituted OCM in a sequence. Figure 35 describes the
scheme for determining the receptors responsible for the responses to two
odorants in a sequence. As shown in Figures 19 to 30, three types of responses
were observed. In the first type of response conductance changes were induced
by both odorants. When DOS induced a conductance change after a
conductance change induced by DES or THP, this was interperted as an
activation of an additional receptor not activated by DES or THP. An alternative
interpertation is that different ligands induce different conductance states and
that the additional conductance induced by DOS is due to the activation of a
different ion channel conductance state(s). The second type of response was a
conductance change produced when DOS was added first and an additional
conductance was not observed when DES or THP followed. Because DOS is
hypothesized to exhibit an osmophore similar to that exhibited by DES and THP,
it was predicted that no additional conductance changes will be observed after
DOS was presented at a saturating concentration. This type of response
supports the prediction but is not a conclusive evidence for the presence of such
osmophore. As brought up above, the reconstituted system is a sample of a
large variety of receptors. It is possible that in these experiments the receptors
for DES and THP were not incorporated into the membrane (if these receptors
indeed do not overlap in their specificity). The third type of response was a
conductance increase induced by DOS, following the absence of any change in
conductance by DES or THP which were added first. This type of response
indicates that DOS does activate a receptor(s) that cannot be activated by DES
or THP.

114

 

 

Start

 

 

I Odorant Tl

No Yes
Odorant B Response 7 Odorant B

 

 

 

 

 

 

No
Response ?

Yes

   

 

Receptor for B l. Receptors for Both A and B —>

but not for A.

No receptors for 1. Receptor for A
A nor B. 2. (a) No receptor for B.

01' A

(b) The receptor for B
is activated by A.

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

—) End

/\

 

 

 

Figure 35. Flowchart of the interpretation of the response of
reconstituted OCM to a pair of odorant. The odorants are introduced one
after the other in saturating concentration.

115

Taken together the three types of response indicate that DOS activates a
receptor that is not activated by DES or THP and possibly that DOS activates
the receptor(s) that DES and THP activate. Another interpretation of one of the
response types is the possibility that DOS induces a different conductance state
than DES or THP by binding to the same receptor(s).

The data presented above suggest that DOS activates an independent
receptor(s), i.e., a receptor(s) that is not activated by DES or THP. The next
question to ask is how independent are the receptors? Specifically, does DOS
activate the receptor(s) activated by DES and THP?

In part of the experiments the odorants were presented one after the other.
The responses can be divided to three types as explained above. Table 9 lists
the response to the six pairs of odorants according to the three response types
(including no response to either odorants, four types). The proportion
(frequencies) of all the experiments that resulted in a response(s) to odorant(s)
is similar to that found in Table 8 although only a part of the experiments is
included in Table 9 (23.1 % vs. 22.0%). The similarity of the proportions indicates
that the subset of the experiments does not deviate from the general trend of the
response proportions.

A response of the reconstituted OCM system to an odorant can be interpreted
as an indication of the availability of a receptor(s) for that particular odorant. A
response however, does not depend only on the presence of the appropriate
receptor in the reconstituted system, but also on the ability of the receptor to
respond to the odorant. A myriad of factors can influence the ability of an
odorant to bind to the receptor and to induce a response of the system. The
reconstitution technique eliminates the mechanisms that involve cytoplasmic-
dependent coupling of the receptors to the ionic channels. Since the

reconstituted system responds to the odorants under similar conditions, the

116

remaining effective factors determining the ability to respond, are those that
affect the binding of the odorant. In the experiments where odorants were
introduced sequentially, the main factor influencing the ability of the system to
respond is the presence of an another odorant. Thus, the availability of the
receptor for a particular odorant depends on the presence of the appropriate
receptor(s) and depends on the inability of the odorant already present to
prevent the binding of the second odorant.

In the context of receptor availability, the responses to odorant pairs should
therefore be divided to two kinds: those which clearly point to the presence of
one receptor and those which can be interpreted in either of two ways. A
response like‘ (-+) is clear, as it indicates absence of a receptor(s) for the first
odorant but a receptor(s) for the second. A response like (+-) can be interpreted
either as absence of a receptor(s) for the second odorant or masking of the
receptor for the second odorant by the first one. The latter interpretation makes
the response (+-) ambiguous for determining the presence of a receptor for the
second odorant. The response (H), as discussed before, can also be
interpreted in more then one way. In the context of this analysis this response
type (++) will be judged as evidence for the presence of two (or more)
independent receptors, one activated by the first odorant and the second
activated by the second odorant.

Table 9 can be used to analyze the independence of the response to each of
the odorants. A response to a particular odorant is independent if the
reconstituted OCM failed to respond to another odorant or if the second odorant

produced an additional response. The independent response proportions are

 

* (+-) indicates a response to the second odorant but not to the first. (-+) indicates a response to
the second odorant but not the first. (++) indicates responses to both first and second odorant.

117

estimates of the probability to find a receptor that will respond only to a particular
odorant but not to the other(s).

Tests that can clearly indicate an independent response to DOS are from the
protocols: DES or THP w 008 (ED stands for 'followed by") where the results
were (-+) or (H). The data show 9 responses in 45 trials (20.0%) (4 + 1 + 3 +1 =
9 out of 27 + 18 = 45). Ambiguous results are from the protocols DOS ®DES or
THP where the results were (+-). The data shows 11 responses of the later type
in 39 experiments (28.2%) (5 + 6 = 11 out of 19 + 20 = 39). Thus, the proportion
of independent response to DOS range from 20.0% to 23.8% (9 I 45 and 20 I84,
The second proportion is when the ambiguous result type is included. (9+11) I
(45 + 39) ).

Test that can clearly indicate an independent response to DES are from the
protocols: DOS or THP Q DES, where the results were (-+) or (++). The data
show no responses in 34 trials (0.0%) (0 + 0 + 0 + 0; 19 + 15). Ambiguous
results are from the protocols DES @0008 or THP where the results were (+-).
The data show 4 responses of the latter type in 46 experiments (8.7%) (2 + 2; 27
+ 19). Thus, the proportion of independent response to DES ranges from 0.0%
to 5.0% (OI 34 and 4 l 80).

Responses that clearly indicate an independent response to THP are from the
protocols: DOS or DES w THP, where the results were (-+) or (H). The data
show no responses in 39 trials (0.0%) (0' + o + o + o; 20 + 19). Ambiguous
results are from the protocols THP $008 or DES where the results were (+-).
The data show 2 responses of the latter type in 33 experiments (6.1%) (0 + 2; 18
+ 15). Thus, the proportion of independent response to THP ranges from 0.0% to
2.8% (0 I 39 and 2 I 72).

118

 

 

 

 

 

Table 11 . Proportions of independent responses to the three odorants.
Proportions
Odorant Minimum 1 Maximum
DES 0.0% 5.0%
THP 0.0% 2.8%
DOS 20.0% 23.8%

 

 

 

 

 

The proportions of responses indicating an independent response to an

odorant (a response not influenced by the presence of another odorant) are

listed in Table 11. The proportions indicate that DOS indeed activated a unique

receptor(s) that DES or THP could not activate. The table suggests another

point: DES or THP did not activate a unique receptor(s). Taken together the data

show that DOS activated a receptor(s) that was not influenced by the presence

of DES or THP and that DES or THP did not activate a receptor(s) in the

presence of DOS.

The following is an illustration of the working model and the interpretation of

Odorants

DES D DOS? THP D—D

Figure 36. A schematic representation of

the three odorants.

the response type frequencies (based
on table 9).

Three odorants were used: DES,
DOS and THP (modeled in Figure 36).
DES is the "core" molecule. DOS is a
di-alcohol derivative of DES. The

additional expected osmophores are

- represented as circles. THP has a

similar chemical formula to that of DES. THP is a cyclic molecule. The

119

hypothetical additional osmophore is represented as a rectangle. Since the
comparison is for the "DES osmophore" (triangle) the osmophore depicted as a
rectangle is expected to have little relevance if any.

Hypothetically, the olfactory cilia may contain four types of putative receptor
types (depicted in Figure 37):
1. A receptor that responds only to DES.
2. A receptor that responds to DES or DOS.
3. A receptor that responds to DOS only.
4. A receptor that responds to DES and DOS (both must be present at the

same time).
Using the model, the next step is to analyze the responses to DES and DOS

using the data in table 9.

 

 

 

 

Figure 38 depicts the
analysis of the response
1 2 3 4 types. The response types
, , , are listed from A to F. A
Figure 37. A schematic representation of the four
WWW" mm"- qualitative analysis gives
the types of receptor that

are needed to be present
in the reconstituted system in order to observe that particular type of response.
Some receptors are crossed off. These are receptors that were obviously absent
if that particular response was observed. Response types C and F require at
least two types of receptors (designated by +) but more then one combination
could support these response types (listed 1 to 4). In the other response types
any of the receptors depicted is sufficient. Responses A, B, D and E point to
some receptor types that were not present in these experiments. The same

responses (A,B,D,E) could be observed if either of the remaining receptor types

120

was present. Therefore, based on conceptual analysis alone, the existence of
any of the receptor types cannot be rejected.

Using the data from table 9 one can assign the frequency that each of the
response types was observed. Table 12 lists the observed frequency of each of
the response types. The table also lists the similar values from the experiments
where THP and DOS were used. Rows D and F in table 12 indicate frequency of
0.0. Thus the probability of incorporating the receptor types 1 or 4 is 0.0.
Eliminating receptor types 1 and 4 allows analysis of the frequencies of the other
two types (types 2 and 3). If receptor types 1 and 4 do not exist, then row A
describes the probability of receptor type 3. Row B is the probability of receptor
type 2 (p1 = 0). Probabilities of 0.0 for receptor types 1 and 4 explains the
frequency of zero of response type F. These probabilities also reduce the
possible combinations of C to only C2 , i.e., receptors 2 and 3. The probabilities
of having two receptors (response C) and observing the same response type
when either receptor is present (response E) should be predicted by the
probabilities of finding each of the receptors.

Before the actual numbers are discussed, here are the laws of compound
probability. In the case of two independent events each of them sufficient for the

outcome, a union of probabilities is used pm, = p, + p3 . In the case of two

independent events that are required to happen at the same time in order for the

outcome to occur, an intersection of probabilities is used pm, = p, x p8 .

121

Table 12. Frequencies of various responses from table 9 arranged
according to the classification in fig 38. Total number of experiments which
tested DES is 27 for A to C and 19 for D to F; Total number of experiments
which tested THP is 18 for A to C and 20 for D to F; Data in the columns are the
number of observations that fit the type of response (A to F) and next to it the
proportion of that response from the experiments it was tested. The pooled
proportions (i.e., probabilities) of the proportions of THP and DES are presented
in the right column (pooled by combining the data from the experiments where
DES was used with those where THP was used).

n - DES p - DES n - THP p - THP Pooled

A 4 0.148 3 0.167 0.156
B 2 0.074 0 0.0 0.044
C 1 0.037 1 0.056 0.044
D 0 0.0 0 0.0 0.0
E 5 0.263 6 0.300 0.282
F 0 0.0 0 0.0 0.0

The probability of finding each of the receptors can be calculated by summing
the frequencies of the responses that indicate the presence of that particular
receptor. The frequencies of responses indicating the presence of receptor type
2 ("DES core") are the sum of the frequency of B and C. The frequency of
responses indicating the presence of receptor type 3 ("DOS extra") is the sum of
the frequency of A and C. Thus the frequency (i.e., probability) of the receptor
type 2 is 0.111 from the experiments using DES and 0.056 from the experiments
using THP. The frequency of the receptor type 3 is 0.185 from the experiments
using DES and 0.223 from the experiments using THP. Response type E should
have a frequency that is the sum of the probabilities of occurrence of receptor

types 2 and 3 (compound probability of independent events). The values from

122

the experiments were DES was used are 0.296 versus 0.263 (calculated vs.
observed). The values from the experiments where THP was used are 0.279
versus 0.30 (calculated vs. observed). Pooling the data from DES and THP
experiments gives values of: 0.088 for receptor type 2, 0.200 for receptor type 3
and 0.288 versus 0.282 for response E (calculated vs. observed). The model
predicts that frequency of response type C is described by the multiple of the
probabilities for receptors 2 and 3. The calculated values are: 0.021, 0.013 and
0.018 (DES, THP and pooled respectively). These values should be compared
to the observed frequencies: 0.037, 0.056 and 0.044 (DES, THP and pooled
respectively). Using the approximate confidence interval (Equation 3) no
significant differences were found between the calculated and the observed
frequencies (p = 99%).

The data obtained from the reconstituted OCM point to two conclusions:

. DOS activates a receptor(s) independent from the one(s) activated by
DES or THP.

. It is unlikely that DOS does not activate the receptor(s) activated by DES
or THP.

The conclusions drawn above are supported by the observation of the
responses to pairs of odorants. The response to DOS after a response to THP or
DES can be interpreted in more then one way. However, the observation that
DOS induced a response after DES or THP did not induce a response, weakens
the alternative mechanism of a single receptor-channel with multiple states.

Thus the data strongly suggest that DOS can activate a response independent
of the one induced by DES or THP.

123

DES —> DOS DOS -> DES

Response Types: Response Types:

A-+E’£EDB D-+EEEB
B+—E]E]EXE( E+—§§(3Q$

 

C ++IEl+Cl F ++i:]+Cj
2 8 + C] 2 G + 8
a E] + B
4 E] + El
Figure 38. Graphic representation of the analysis of the types of response

in terms of the required receptors.

124

The next question to ask is whether DOS also activates the receptor(s) that
mediates the response to DES or THP? The evidence to answer this question is
not as direct as for the first observation. If the response(s) of the system to DOS
is truly unrelated to that induced by DES or THP, then the frequencies of the
responses to DES and THP should not be influenced by the presence of DOS.
Looking at the frequencies of response types shows that this is clearly not the
case. A response to either DES or THP has never been observed when DOS
was present in the system when they were introduced. Furthermore the
frequency of response to DOS when it was the second odorant together with the
frequency of responses to DES or THP when they were the first odorant roughly
matches the observed frequency of responses to DOS when it was the first
odorant. This match, together with the lack of response to DES or THP in the
presence of DOS suggest the DOS binds to the same receptor(s) that DES or
THP do. The data do not show if DOS actually activates this (these) receptors.
On the basis of the structure of the odorants it is likely that DOS does activate it
but no direct evidence supports this hypothesis. Minute structural differences
may cause DOS to be an antagonist rather than an agonist.

To provide corroborating evidence for the in vitro observation an in vivo
measurement of the competitive binding-was performed. The mixture
discrimination index (Equation 1) was calculated for DOS and DES. The MDI
was found to be 1.43, a value which is interpreted in the literature as a partial
discrimination of the odorants by the receptors. This means that the some of
receptors present in the OE cannot discriminate between the odorants but others
do.

The addition of the data from the in vivo experiments to the observation made
on the results of reconstitution experiments suggests that DOS does activate

the receptor(s) activated by DES or THP. In the framework of the multiple profile-

125

multiple receptor site, the evidence obtained in this study suggests that DOS is
composed of at least two osmophores: one is similar to that of DES and THP

and the other is different.

CONCLUSIONS

. Dose-response relationships of reconstituted olfactory cilia membrane and
the odorants: diethylsulfide, diethanolsulfide and thiophene, were established
. The effective doses of diethylsulfide were found to be: ED5 = 1.05 nM, ED50
= 2.63 nM and ED95 = 7.59 nM.

. The effective doses of diethanolsulfide were found to be: ED5 = 83.2 nM,
E050 = 126 nM and ED95 = 195 nM.

. The effective doses of thiophene were found to be: ED5 = 0.28 nM, ED50 =
0.71 nM and ED95 = 2.16 nM.

. The effects of the odorants did not require ATP or GTP, thus suggesting that
the transduction of these odorants Is not mediated by second messengers.

. Diethanolsulfide induced conductance changes independent from those
induced by diethylsulfide or thiophene.

. Diethylsulfide or thiophene did not induce conductance changes independent
of those induced by diethanolsulfide.

. The observation that only diethanolsulfide induced conductance changes
independent of the two other odorants suggests that this odorant binds to the
same receptors as the other two odorants and an additional receptor(s)
which the two other odorants do not bind to.

. Mixture discrimination index was calculated based on EOG recordings. The
value of the index for the mixture of diethylsulfide and diethanolsulfide was

1.43.

126

127

10. The value of mixture discrimination index is interpreted as a partial overlap
between the receptors that diethanolsulfide can activate and those activated
by diethylsulfide.

11.Taken together, the results of the reconstitution experiments and the mixture
discrimination index support the hypothesis that diethanolsulfide presents to
the olfactory receptors a substructure similar to that presented by
diethylsulfide and thiophene and an additional substructure. The two

substructures bind to different receptors.

APPENDICES

APPENDIX A

Acetone Extraction Procedure

SBTPE was dissolved in ether (33 gm in 90 ml) at room temperature under
argon. The SBTPE in ether solution was added to acetone using a small pipette
(350 ml) while stirring vigorously. The suspension initially appeared yellow-white
and became brown-orange with the increase of lipid concentration. After stirring
for 1 hour, the suspension was allowed to settle for 30 minutes. The suspension
separated to a brown precipitate and a yellow supernatant. The supernatant was
removed and acetone was added to the precipitate (500 ml). The suspension
was then stirred constantly for 24 hours. After 24 hours, the suspension was
allowed to settle again. The suspension separated to a yellow-light brown
precipitate and a light yellow supernatant. The supernatant was removed. The
precipitate in a form of paste was dried under vacuum. The dried lipids were
dissolved in chloroform (as little volume as possible) and spread on clean glass
to dry, scraped off the glass, dried under vacuum and weighed. The yield was
about 50%. The extracted lipids were dissolved in chloroform (10% wlv) and

stored in a freezer (-20°c). The lipid composition was not determined.

128

APPENDIX B

Layout of the connection box printed circuit board

 

e 0 A81
Vin R1
. ///Sg

° 0’0
V C Elect
n e e e

V Shields

Vin - Voltage input; R1 - Resistor (10m); I - Picoammemeter connection. The
picoammeter is connected floating; V - Voltmeter connection (single ended); C -
Capacitance meter connection; 81 - Switch (SPST); 82 - Switch (SPDT);

 

 

 

129

APPENDIX C

I-V measurment program
Code Files:

I‘
ivmain.cpp
*I

#include'lvh"
#include'configh"
#include'logo.h"
#include<string.h>
#include<conio.h>

maino
{
int gdriver = DETECT, gmode,t_error;
char ans = O;
struct textsettingstype texttypeinfo;
Config *SysConfig;
float a,b;
IV Task;
#ifndef MOCK
Voltage *v;
Current ‘0;
Logo *l;
K570 sysz

v = new Voltage(&sys);
c = new Current(22);
c->set_remote0;

130

131

o->set_auto();

Task.V = v;
Task.l = 0;

#endif

l = new Logo;
delete I;

SysConfig = new Config;

do
{

SysConfig->review0;

II Copy the configuration parmeters

Task.stepTime = SysConfig->ivT I 5; ll Each tick is 5 mSec.
Task.delayTime = SysConfig->ivD I 5;
Task.interval = SysConfig->ivl I 5;

Task.nMeasunnents = SysConfig->ivNM;
Task.nOfSteps = SysConfig->ivNS;

Task.MinV = SysConfig->ivMinV;
Task.MaxV = SysConfig->ivMaxV;
strcpy(Task.fn,SysConfig->fn);
Task.minX = SysConfig->minX;
Task.maxx = SysConfig->maxx;
Task.minY = SysConfig->minY;
Task.maxY = SysConfig->maxY;

b = SysConfig->vMult;

a = SysConfig->vOffset;

clrscro;
printf("\nDo you want to calibrate? (y/ n).. "); ans = getcheo;
if(ans == 'y')

I

v->calibrate(a,b);

SysConfig->vMult = b;
SysConfig->vOffset = a;

printf("\nAlpha is: %f Beta is: %fln",a,b);
WOW:

I;

v->changeCF(b,a);

iniLinkedGraphies(&gdriver,&gmode); .
registerbgifont(DEFAULT_FONT);
I‘

initGraphim(&gdriver,&gmode,"c:\\bor1andc\\bgi');
*I
settextstyle(DEFAULT_FONT,HORIZ_DIR,1);

132

Task.makeScreeno;

Task.mnTasko;
closegraphO:
printf('Any more? yIn\n"):
ans = getcheo;
if((ans == 'y') H (ans == 'Y')) ans = 1;
else ans = 0:
}while(ans);
delete SysConfig;
return 0;
}
It
IV.CPP
*I
#include"iv.h"

II #include'usen'o.h'
#include<conio.h>
#include<stdio.h>
#include<alloc.h>
#include<time.h>
#include<stdlib.h>

extem volatile long S_CLOCKS[4];

void lV::makeScreen(void)

{
struct Video videoMax;
getVideoMax(&videoMax);
II Graph window //

graph = new Gwindow (videoMax.maxx * 0.15 ,0,
videoMax.maxx, videoMax.maxy * 0.85 -1,
shown,0,VGA_BLUE,VGA_BLUE,VGA_DARKGRAY,VGA_WHITE,
1.0.1.
minX,minY,maxX,maxY);

II Text windows I/
imax = new PlainTextth (0,0, videoMax.maxx * 0.15 -1 , videoMax.maxy * 0.05-1,

'”,shown,0,
VGA_BLUE,VGA_CYAN);

133

imin = new PlainTextWin (0, videoMax.maxy * 0.80,
videoMax.maxx ' 0.15 -1 , videoMax.maxy ' 0.85-1,
"",shown,0,
VGA_BLUE,VGA_CYAN);

ititle = new PlainTextWin (videoMax.maxx * 0.10, videoMax.maxy * 0.05,
videoMax.maxx ‘ 0.15 -1, videoMax.maxy " 0.80-1,
'Current",shown,0,
VGA_BLUE,VGA_CYAN,
VERT_DIR,CENTER_TEXT.CENTER_TEXT,0.5F,0.5F,
DEFAULT_FONT,2);

vmin = new PlainTextWin (videoMax.maxx * 0.15, videoMax.maxy ' 0.85,
videoMax.maxx * 0.30 -1 , videoMax.maxy * 0.90 -1,
"",shown,0,
VGA_BLUE,VGA_CYAN,
HORIZ_DIR,LEFT_TEXT,CENTER_TEXT,0.0F,0.5F);

vmax = new PlainTextWin (videoMax.maxx * 0.85, videoMax.maxy * 0.85,
videoMax.maxx, videoMax.maxy * 0.90 -1,
"",shown,0,
VGA_BLUE,VGA_CYAN,
HORlZ_DlR,RIGHT_TEXT,CENTER_TEXT,1 .0F,0.5F);

vtitle = new PlainTextWin (videoMax.maxx * 0.30, videoMax.maxy " 0.85,
videoMax.maxx * 0.85 -1, videoMax.maxy " 0.90 -1,
”Voltage",shown,0,

VGA_BLUE,VGA_CYAN,
HORIZ_DIR,CENTER_TEXT,CENTER_TEXT,0.5F,0.5F,
DEFAULT_FONT,2);

nSteps = new PlainTextWin (videoMax.maxx * 0.16, videoMax.maxy * 0.91,
videoMax.maxx " 0.35 -1 , videoMax.maxy,
“Step #:”,shown,0,VGA_BLACK,VGA_LIGHTRED,
I-IORIZ_DIR,CENTER_TEXT.CENTER_TEXT,0.5F,0.5F);

fileName = new PlainTextWin (videoMax.maxx * 0.36. videoMax.maxy * 0.91,
videoMax.maxx * 0.84 -1 , videoMax.maxy.
",shown,0,VGA_BLACK,VGA_LIGHTRED,
HORIZ_DIR.CENTER_TEXT,CENTER_TEXT,0.5F,0.5F);

 

I**“"“ Buttons I
mnswitch = new TextButton (0, videoMax.maxy " 0.86,
videoMax.maxx " 0.14, videoMax.maxy,
'RUN',"STOP”,VGA_WHITE, VGA_BLACK,VGA_RED,VGA_GREEN,
HORIZ_DIR,CENTER_TEXT,CENTER_TEXT,0.5F,0.5F,
DEFAULT_FONT,2);

status = new TextButton (videoMax.maxx ' 0.85 -1, videoMax.maxy ' 0.91,
videoMax.maxx, videoMax.maxy,
'Measuring','Waiting');

 

I‘“‘*"‘ PopUp I

void

134

msg = new PlainTextVtfin (videoMax.maxx * 0.35,videoMax.maxy * 0.45,
videoMax.maxx * 0.65 , videoMax.maxy * 0.57,
"The IV will reset...",hidden,1,
VGA_WHITE,VGA_RED);

I‘m‘ Graph limits WI

sprintf(vmin->text,"%6.3G',minX);
sprintf(vmax->text,"%6.36",maxX);
sprintf(imin->text,"%6.3G”,minY);
sprintf(imax->text,"%6.3G",maxY);
sprintf(fileName->text,'Current file: %s",fn);
vmin->putText0;

vmax->putText0;

imin->putTexI0;

imax->putText0;

fileName->putText0;

sprintf(status->inActiveText,"Idle");
status->inactivateo;
graph->drawLine(minX,0.0F,maxX,0.0F,VGA_CYAN);
graph->drawLine(0.0F,minY,0.0F,maxY,VGA_CYAN);

lV::runTask(void)

II Initialize loop variables

int ans = 0, ui = 0;
int running = 0;
int i = 0;

int lastN = 0;
FILE *datafile;

table = (point ‘)calloc(nOfSteps * nMeasurments,sizeof(point»;
temp = (point ‘)calloc(nMeasurments,sizeof(point));

datafile = fopen(fn."wt"):

dV = (MaxV - WW) I (nOfSteps -1):
step = 0;

x=y=00fi

xo = yo = 0.0F;
sprintf(nSteps->text,"Step #: %d",step);
nSteps->clearWin0;
nSteps->putText0;

do

{
if((S_CLOCKS[0] <= cuss running) // If it is time.

135

{
S_CLOCKS[0] = stepTime; II Reset the timer
if(ste == 0)

{

V->setVoltage(MinV + dV * (float)step);

S_CLOCKS[0] = stepTime * (nOfSteps - 1); II Reset the timer
while(S_CLOCKS[0] > 0) ;

S_CLOCKS[0] = stepTime; II Reset the timer
I:
measurmento;
for(t=0;i<nMeasurments;i++)

{
drawCross(temp[i][0],temp[i][1],VGA_RED);
table[i+lastN][0] = temp[i][O];
table[i+lastN][1] = temp[i][1];

};

lastN += nMeasurrnents;

step“;

if(step > 1)
graph->drawLine(xo,yo,x,y,VGA_WHITE);

sprintf(nSteps->text,"Step #: %d",step);

nSteps>clearWin0;

nSteps->putText0;

xo=x;
Y°=Y;

};

/I Check the KBD. If hit get the key, if not set 'ans' to 0.

if(kbhitO) ans = getchO;
else ans = 0;

II Process the character entered.

switch(ans)

I

case '3': II Stop

case 'S':
runswitch->inactivate();
sprintf(status->inActiveText,"ldle");
status->inactivate0;
if(running) reset_clock0;
running = 0;
break;

case 'r': II Run

case 'R':

runswitch->activate();
sptintf(status->inActiveText,”Waiting');
status->inactivate0;

start_clock();

S_CLOCKS[0] = 0L;

running = 1;

break;

case 'q':
case '0':

case 'i':
case 'l':

case '0':

136
// Quit

if(running) reset_clock0;
running = 0;

ans = -2;

break;

i=0;

II Interrupt. PopUp messege window

if(running)

case 'a':
case 'A':

reset_clock0; II Stop the clock
runswitch->inactivate0;

running = 0;
graph->clearWin0;
x = y = 0.0F; II Reset values
xo = yo = 0.0F;
step = 0;
i =1;
};
ui = POPUPO:
If(i)
{
runswitch->activate(); II Restart
start_clocko;
S_CLOCKS[0] = stepTime;
running = 1;
};
break;
I/ Apply 0.
I = 0;
if(running)
reset_clock0; II Stop the clock
runswitch->inactivate0;
running = 0;
graph->clearWin0;
x = y = 0.0F; II Reset values
x0 = yo = 0.0F;
step = 0;
i =1;
};
V->setVoItage(0.0F);
break;
I/ Apply reconstitution V
i = 0;
if(running)

reset_clock0; // Stop the clock
mnswitch->inactivate();
running = 0;
graph->clearWin0;

137

x = y = 0.0F; II Reset values
x0 = yo = 0.0F;
step = 0;

i=1;

i:
adjustVoltageO;

break;
case 't': II Apply thinningV
case 'T':

i= 0;

if(running)

I

reset_clock0; l/ Stop the clock
mnswitch->inactivate();
running = 0;
graph->clearWin0;
x = y = 0.0F; II Reset values
xo = yo = 0.0F;
step = 0;
i =1;

I;

V->setVoItage(-0.1 5F);

break;

}:
}while((ans >= 0) 88 (step < nOfSteps));

mnswitch->inactivate();
if(running) reset_clock0;
running = 0;

sprintf(status->inActiveText,"Done");
status->inactivate0;
V->setVoltage(0.0F);

if(ans >= 0)
for(i=0;i<lastN; i++)
fprintf(datafile,"%10.3G , %10.3G\n",table[i][0],table[i][1]);
fclose(datafile);

free(table); free(temp):

while(lkbhit());
};
void IV::measurment(void)
{ int i;

float r,

II Actual timed task.
#ifndef MOCK

V->setVoItage(MinV + dV " (float)step); II Set the voltage

138

#endif
S_CLOCKS[1] = delayTime; II Wait for the delay
while(S_CLOCKS[1] > 0L) if(kbhitO) return;
status~>activate();
for(i= 0;i<nMeasunnents;i++)
#ifdef MOCK
randomizeo;
r = (float)(rand0 - RAND_MAX I2) I (float)RAND_MAX "' 2;
temp[i][O] = MinV + dV ' ((float)step) + r " 0.01;
temp[i][1] = temp[i][O] " 2.55-9;
#else
V->getVoltage(temp[i][0]);
l->getCurrent(temp[i][1]);
#endif
S_CLOCKS[1] = interval; II Wait between measurrnents

int

void

while(S_CLOCKS[1] > 0L) if(kbhitO) return;
status->i’nactivate0;

x = 0; l/ Average
Y = 0;

for(i= 0;i<nMeasurments;i++){x += temp[i][O]; y+=temp[i][1];};

x I= (float)nMeasunnents; '

y I= (float)nMeasunnents;

table[step][0] = x;

tablelstepllll = y:

lV::popUp(void)

int i;
if(lmsg->stored) II Display messege
I

msg->openWindow0;
msg->putText0;

I
else msg->showWindow0;
oetcho;

msg->hideWindow0;
retum(i);

lV::drawCross(float x, float y, int color) -

float tX,tY;

tX = (maxX - minX) " 0.01;
W = (maxY - minY) * 0.01;

graph->drawLine(x,y+tY.x.y-IY.COIOT):

139

graph->drawLine(x+tX,y.x-tx.y,color);
I;

void IV::adjustVoltage(void)

int ans = 0;

float v = 0.0F;
do
ans = getcho;
if(ans == 0) ans = getcho; // Extended code
switch(ans)
{
case 72: IIUP arrow
v+=0.001;
break;
case 80: I/ Down Arrow
v -= 0.001;
break;
case 13: I/ ENTER
ans = 0;
break;
I;
V->setVoItage(v);
}while(ans l= 0);
It
Config.cpp
*/
#include'\cpp\k570\config.h"

#include<string.h>
#include<conio.h>
#include<stdlib.h>

{Config::Config(char 50)

II Open file
f=fopen(s,"rt");
if(lf) return;
else {
readConfigo;

strcpy(cfn.8);

I
nB = BLUE; nF = WHITE; hF = BLUE; h8 = LIGHTGRAY;
col = 35;
min = 1 ;max = 22;

140

void Config::readConfig(void)

{

char buf[200],st[10];
fread(buf, 200, 1. 0:
char‘ p;

p = strtok(buf,"\n"); adR = atoi(p);
p = strtok(NULL,"\n"); daR = atoi(p);

p = strtok(NULL,"\n" ); vMult = atof(p);

p = strtok(NULL,"\n" ); vOffset = atof(p);

p = strtok(NULL,"\n" ); iMult = atof(p);

p = strtok(NULL,"\n" ); iOffset = atof(p);

p = strtok(NULL,'\n" ); cVO = atof(p);

p = strtok(NULL,"\n" ); chesduaI = atof(p);
p = strtok(NULL,"\n" ); 005 = atof(p);

P = stfl0k(NULL.'\n" ): strepy(fn.p );

p = strtok(NULL,"\n" ); ivT = atol(p);

p = strtok(NULL,'\n" ); ivD = atol(p);

p = strtok(NULL,"\n" ); ivl = atol(p);

p =strtok(NULL,"\n" ); ivNM = atoi(p);

p =strtok(NULL,"\n" ); ivNS = atoi(p);

p = strtok(NULL,"\n" ); ivMinV = atof(p);
p = strtok(NULL,"\n" ); ivMaxV = atof(p);
p = strtok(NULL,'\n" ); ivMV = atof(p);
p = strtok(NULL,"\n" ); minx = atof(p);
p = strtok(NULL,"\n" ); maxX = atof(p);
p = strtok(NULL."\n” ); minY = atof(p);
p = strtok(NULL,"\n" ); maxY = atof(p);

Config::~Config(void)

void

fclose(f);

Config: :saveConfig(char s[])

FILE' oF;

if(lstrcmp(cfn,s)) fclose(f);

oF = fopen(s,"wt');

fptintf(oF,"%d\n",adR); // 1
fprintf(oF,"%d\n",daR); II 2
fprintf(oF,"%g\n",vMult); // 3
fpiintf(oF,"%g\n",vOffset); II 4
fprintf(oF,"%g\n”,iMult); ll 5
fptintf(oF,"%g\n",iOffset); II 6
fprintf(oF,"%g\n",cV0); II 7

fprintf(oF."%g\n",chesduaI); /I8
fprintf(oF,"%g\n",cCs); /I 9

void

{

It

141

fprintf(oF,"%s\n",fn); // 10
fprintf(oF,"%ld\n",ivT); I/ 11
fprintf(oF,"%ld\n",ivD); II 12
fprintf(oF,"%ld\n",ivl); II 13
fprintf(oF,"%d\n",ivNM); 'II 14
fprintf(oF,"%d\n",ivNS); II 15

fprintf(oF,"%g\n",ivMinV); II 16
fprintf(oF,"%g\n",ivMaxV); II 17

fprintf(oF,"%g\n",ivMV); I/ 18
fprintf(oF,"%g\n",minX); I/ 19
fprintf(oF,"%g\n",maxX); II 20
fprintf(oF,"%g\n",minY); II 21
fprintf(oF."%g\n",maxY); II 22

Config::review(void)

int i,oldi;
int ans;

/I Set video
textbackground(nB);
textcolor(nF);
clrscrO;

Display data *I

II Write Iables

cprintf("T he AD range 1-5\n\r');
cprintf("T he DA range 1-5\n\r");
cprintf("V multipliennV');

cprintf("V offset\n\r");

cprintf("l MultipliennIr“);

cprintf(“l Offset\n\r");
cprintf("Capacitace V0\n\r");
cprintf(“Capacitance V residual\n\r");
cprintf(”Capacitance Cs\n\r");
cprintf(”Default data file name\n\r");
cprintf("Total step time (mSec)\n\r");
cprintf(”Delay to measurrnents (mSec)\n\r");
cprintf("lnterval between measurments\n\r");
cprintf(‘# of Measurments\n\r");
cprintf("# of steps\n\r");
cprintf(”Minimun V\n\r");
cprintf(”Maximum V\n\r");

cprintf(”Max +I- voltage (Volts)\n\r");
cprintf("Minimun X (plot window)\n\r");
cprintf("Maximum X (plot window)\n\r“);-
cprintf(”Minimum Y (plot window)\n\r");
cprintf(”Maximum Y (plot window)\n\r");

1 42
cprintf(‘\n\r\n\rENTER to change ESC to leave");

II Write values

gotoxy(col,1); cprintf("%d",adR);
gotoxy(col,2); cprintf("%d",daR);
gotoxy(col,3); cprintf(”%g",vMuIt);
gotoxy(col,4); cprintf("%g",vOffset);
gotoxy(col,5); cprintf(”%g",iMult);
gotoxy(col,6); cprintf("%g",iOffset);
gotoxy(col,7); cprintf("%g",cV0);
gotoxy(col,8); cprintf('%g”,chesdual);
gotoxy(col,9); cprintf(“%g”,cCs);
gotoxy(col,10); cprintf(”%s",fn);
gotoxy(col,11); cprintf("%ld",ivT);
gotoxy(col,12); cprintf("%ld",ivD);
gotoxy(col,13); cprintf(”%ld",ivl);
gotoxy(col,14); cprintf(”%d",ivNM);
gotoxy(col,15); cprintf("%d",ivNS);
gotoxy(col,16); cprintf(”%g",ivMinV);
gotoxy(col,17); cprintf(”%g”,ivMaxV);
gotoxy(col,18); cprintf("%g",ivMV);
gotoxy(col,19); cprintf(”%g\n”,minX);
gotoxy(col,20); cprintf(”%g\n",maxX);
gotoxy(col,21); cprintf(”%g\n",minY);
gotoxy(col,22); cprintf(”%g\n",maxY);

II Loop
i = oldi = 1;
hiLight(i);

ans = 0;

do

I
ans = getcho;
if(ans == 0) ans = getcho; II Extended code
switch(ans)

I
case 77: II Right Key
case 72: II Up Key
|-;
if(i<min) i = max;
gotoxy(col.i);
loLight(oIdi);
hiLight(i);
gotoxy(col,i);
oldi = i;
break;
case 75: II Right Key
case 80: II Up Key
i++;
if(i>max) i = min;
gotoxy(col,i);
loLight(oIdi);
hiLight(i);
gotoxy(col,i);
oldi = i;

143

break;

case 13:
gotoxy(col,i);
clreolo;
getVar(i);
loLight(i);
gotoxy(col,i);
break;

case 27:
ans = -1;
break;

II Enter

// ESC

I
}while(ans > 0);

return;

I:

void

{

Config::hiLIght(int i)

gotoxy(col,i);
clreolo;
textbackground(nB);
textcolor(hF);
gotoxy(col,i);
switch(i)
I
case 1:
case 2:
case 3:
case 4:
case 5:
case 6:
case 7:
case 8:
case 9:

cprintf(‘%d",adR); break;
cprintf(“%d“,daR); break;
cprintf(”%g",vMult); break;
cpnntf(‘%g',vOffset); break;
cprintf("%g",iMult); break;
cprintf("%g'JOffset); break;
cprintf("%g",cV0); break;
cprintf(”%g",chesdual); break;
cprintf(”%g",cCs); break;

case 10:
case 11:
case 12:
case 13:
case 14:
case 15:
case 16:
case 17:
case 18:
case 19:
case 20:
case 21:
case 22:

cprintf(”%s",fn); break;
cprintf(”%ld",ivT); break;
cprintf(”%ld",ivD); break;
cprintf("°/old",ivl); break;
cprintf(”%d",ivNM); break;
cprintf(”%d”,ivNS); break;
cprintf(”%g",ivMinV); break;
cprintf("%g",ivMaxV); break;
cprintf(”%g”,ivMV); break;
cprintf(”%g\n",minX); break;
cprintf(”%g\n",maxX); break;
cprintf("%gIn'minY); break;
cprintf("%g\n",maxY); break;

I:
textbackground(nB);
textcolor(nF);

return;

void

return;

I;

void

144

Config::loLIght(int i)

9010XY(00|.i):
clreolo;
textbackground(nB);
textcolor(nF);
gotoxy(col,i);
switch(i)

I

case 1: cprintf("%d',adR); break;
case 2: cprintf(”%d",daR); break;
case 3: cprintf(“%g",vMult); break;
case 4: cprintf(”%g",vOffset); break;
case 5: cprintf(”%g'JMult); break;
case 6: cprintf("%g",iOffset); break;
case 7: cprintf(“%g",cV0); break;
case 8: cprintf("%g",chesduaI); break;
case 9: cprintf(”%g",cCs); break;
case 10: cprintf("%s”,fn); break;

case 11: cprintf("%ld",ivT); break;
case 12: cprintf(“%ld'JvD); break;
case 13: cpnntf('%ld',ivl); break;
case 14: cprintf(”%d",ivNM); break;
case 15: cprintf("%d",ivNS); break;
case 16: cprintf(”%g”,ivMinV); break;
case 17: cprintf(”%g”,ivMaxV); break;
case 18: cprintf(”%g",ivMV); break;
case 19: cprintf(”%g\n",minX); break;
case 20: cprintf(“%g\n",maxX); break;
case 21: cprintf(“%g\n",minY); break;
case 22: cprintf(“%g\n",maxY); break;

Config::getVar(int i)

gotoxy(col,i);
clreolo;
textbackground(nB);
textcolor(hF);
gotoxy(col,i);
switch(i)

I

case 1: cscanf("%d",&adR); break;
case 2: cscanf("%d",&daR); break;
case 3: cscanf("%g",&vMult); break;
case 4: cscanf(‘%g",&vOffset); break;

I.“

 

145
case 5: cscanf('%g",&iMult); break;
case 6: cscanf("%g",&iOffset); break;
case 7: cscanf("%g",&cV0); break;
case 8: cscanf("%g",&chesdual); break;
case 9: cscanf("%g",&cCs); break;

case 10:
case 11:
case 12:
case 13:
case 14:
case 15:
case 16:
case 17:
case 18:
case 19:
case 20:

cscanf("%s",&fn); break;
cscanf("%ld",&ivT); break;
cscanf(‘%ld",&ivD); break;
escanf('%ld",&ivl); break;
cscanf("%d",&ivNM); break;
cscanf("%d",&ivNS); break;
escanf(”%g",&ivMinV); break;
cscanf("%g",&ivMaxV); break;
cscanf("%g",&ivMV); break;
cscanf(”%g\n",&minX); break;
cscanf(‘%g\n",&maxX); break;

case 21: cscanf("%g\n",&minY); break;
case 22: cscanf("%g\n",&maxY); break;
I:
textbackground(nB);
textcolor(nF);
return;
I;
It
G_WIND.CPP
Implantation of G_WIND.H
*I

#include<alloc.h>
#include<graphics.h>
#include'\cpp\k570igwind\g_wind.h'

Gwindow::Gwindow(int l, int u, int r, int b, II Viewport
enum visibility see, II Open shown or hidden.
int savebg, I/ 0 = no 1 = yes (def 0);
int cbg, II Bg color.
int cbbg, II Border bg color.
int cbfg, // Border foreground color.
int ca, 'II Active forgraound color.
int clp, II On.
int border, II No border.
int dir, II 1 = real (def 1).
float xmin, II Dimensions. (def 1X1).
float ymin,
float xmax,
float ymax
I

W Initialize object intemal variables ********I

 

146

if(lsavebg) hideCapable = 0;
else hideCapable = 1;

II Status variables

visible = hidden;
saved = 0;
stored = 0;

clip = clp:
II Set viewport

totallefl = I;
total.up = u;
total.right = f;
total.bottom = b;
totaI.x = r-l+1;
total.y = b-u+1;

II Set colors and fill background;

colors.bg = cbg;
colors.bbg = cbbg;
colors.bfg = cbfg;
colors.current = ca;

// Calculate the net viewport for the window

switch(borderType)

I

case 1:
net.left = I + 2;
net.up = u + 2;
net.right = r - 2;
net.bottom = b - 2;
break;

case 2:

net.left = l + 4;
net.up = u + 4;
net.right = r - 4;
net.bottom = b - 4;
break;
default:
net.left = I;
net.up = u;
net.right = r,
net.bottom = b;
I:
borderType = border,
xpixels = net.right - net.left;
ypixels = net.bottom - net.up;

I/ calculate conversion ratios.

 

147

dim.minx = xmin;
dim.miny = ymin;
dim.maxx = xmax;
dim.maxy = ymax;
dim.directions = dir;

xratio = (xmax - xmin) I float(xpixels);
yratio = (ymax - ymin) I float(ypixels);

if((see == shown) || lhideCapable) openWindowo;

 

 

I;
Gwindow::~Gwindow(void)
I
if(saved)
farfree(bgimage);
if(stored)
farfree(storedimage);
I:
I"m Generalwindowmaintance“”””““‘””‘““““”“I

 

II Clear the working area. (Public)
void Gwindow::clearWIn(void)

setfillstyle(SOLID_FILL.colors.bg);
setNeto;
bar(0,0,xpixels,ypixels);

// Draw the background and border. (Protected)
void Gwindow::drawWin(void)
if(bonderType)

setfillstyle(SOLlD_FlLL,colors.bbg);

setTotalo;

bar(0,0,total.right-total.Ieft,total.bottom-total.up);

setcolor(colors.bfg);

rectangle(0,0,total.right-total.left,total.bottom-total.up);

if(borderType == 2)
rectangle(2,2,total.right-total.left-2,total.bottom-total.up-2);

setNeto;
cleartho;

148

II Open the window. This has to be the first call to a window
II that was open hidden. (Public)

void Gwindow::openWindowo

I
unsigned size;
if(hideCapable 88. lvisible) II Store bg only if hide capable
I
saved = 0;
II Get the bg image
setTotalo;
size = imagesize(0,0,total.x,total.y);
if(size < 0xFFFF)
I
bgimage = farmalloc(size);
if(bgimage != NULL)
I
getimage(0,0,total.x.total.y,bgimage);
saved = 1;
I;
I:
I:
if(lvisible) II Draw only if not visible
I
drawWInO;
visible = shown;
I:
I:

II Close a visible or stored window freeing the memory. No calls to
II the window should be preformed without a call to openWin. (Public)

void Gwindow::closeIMndow(void)

if(saved)

I
setTotalo;
putimage(0,0,bgimage,COPY_PUT);
farfree(bgimage);
saved = 0;
visible = hidden;
I:
if(stored)

I

setTotalo; .
putimage(0,0,storedimage,COPY_PUT);
farfree(storedimage);

stored = 0;

visible = shown;

I:

149

/I Hide a visible window (Public)

void Gwindow::hideWindow(void)

{

unsigned size;
if(hideCapable 88. visible)

I

[I Get the window image

setTotalo;

size = imagesize(0,0,total.x,total.y);
if(size < OxFFFF)

storedimage = fannalloc(size);
if(storedimage l= NULL)

I
getimage(0,0,total.x,total.y,storedimage);
stored = 1;

I;

/I Put the background.
setTotalo;
putimage(0,0,bgimage,COPY_PUT);
visible = hidden;
saved = 0;
farfree(bgimage);

II Show a hidden window. (Public)

void Gwindow::showWindow(void)
I

unsigned size;
if(stored 8.8. lvisible)

I/ Get the bg image
setTotalo;
size = imagesize(0,0,total.x,total.y);
if(size < OxFFFF)
I
bgimage = farmalloc(size);
if(bgimage I= NULL)
I

getimage(0,0,total.x,total.y,bgimage);
saved = 1;

I:

II Put the stored window.
setTotalo;
putimage(0,0,storedimage,COPY_PUT);

150

 

 

visible = shown; II Visible.
stored = 0; II Not stored.
farfree(storedimage); II Free the window image memory.
I:
I;
I:
I AAAAAAAAAA ; “““““““““““““““““““““““““““““““““ I
I'““ Window arithmetics “““““““““““““““““““““““ I

 

II Set viewport to working area. (Private)

void Gwindow::setNet(void)

I
I;

setviewport(net.left,net.up,net.right,net.bottom,1);

II Set viewport to total area. (Private)

void Gwindow::setTotaI(void)

I
I:

setviewport(total.left.total.up,total.right,total.bottom,1);

II Convert x real coordinate to a int x in the viewport. (Protected)

int Gwindow::xToi(float x)

{

int i;

i = (x - dim.minx) I xratio;
retum(i);

I

// Convert y real coordinate to a int y In the viewport. (Protected)

int Gwindow::yToj(float y)
I

int i;

i = (y - dim.miny) / yratio;
if(dim.directions)
i = ypixels - i;
retum(i);

int Gwindow::setDimensions(float xmin, float ymin, float xmax, float ymax)

I

151

dim.minx = xmin;
dim.maxx = xmax;
dim.miny = ymin;
dim.maxy = ymax;

xratio = (xmax - xmin) I float(xpixels);
yratio = (ymax - ymin) I float(ypixels);
retum(O);
I

AAAAAAAAA‘AA‘AAA-AAALA‘AAAA‘AAAJAAAAAA-AAALAAAAAAAAAA/

A-AAA‘A
PW w. I I . AA‘AAAA‘AAA‘AA-AA AAA‘AAAAALAA-L/
-AAAA‘A-AAALA‘LAAA‘AA-A AAA AAAA

‘AAAA AAAAAAAA‘AAAAAAAAAAAAAAALLAA‘LAA‘AAA‘ I

 

 

II From (x1,y1) to (x2,y2) using color 'color‘.

int Gwindow::drawLine(float x1, float y1, float x2, float y2, int color)
I

int a,b,c,d;

 

 

setcolor(color);
setNeto;
a = xToi(x1);
c = xToi(x2);
b = yTol(y1):
d = yTOIIyZ):
line(a,b,c,d);
retum(O);
I:
// Color pixel at (x,y).
int Gwindow::drawPoint(float x, float y, int color)
I
int a,b;
setNetO;
a = xToi(x);
b = yToij);
putpixel(a,b,color);
retum(O);
I;
I'“” Window cursor interface I
I“ AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA /

 

II Are the physical screen coordinates (x,y) in the window?

int Gwindow::isanIfinflnt x, int y)
I

152

int resultx = 0, resulty = 0. result = 0;

if((x >= total.left) 88 (x <= total.right» resultx = 1;
if((y >= total.up) 88 (x <= total.bottom» resulty = 1;
if(resultx 88 resulty) result = 1;

retum(result);

I.

/I Are the physical screen coordinates (x,y) in the work area?

int Gwindow::isanorkArea(int x, int y)

I
int resultx = 0, resulty = 0. result = 0;
if((x >= net.lefl) 88 (x <= net.right» resultx = 1;
if((y >= net.up) 88 (x <= net.bottom» resulty = 1;
if(resultx 88 resulty) result = 1;

retum(result);

I;

II Translate the physical coordinates to real window coordinates.

int Gwindow::wherelsCrs(int x, int y, struct Rcoord ‘r)

I
int flag;
float w,z;

flag = isInWorkArea(x,y);

if(flao)
w = (float)(x - net.right) * xratio + dim.minx;
z = (float)(y - net.up) " yratio;
if(dim.directions)
z = dim.maxy - 2;
else
2 = dim.miny + z;
I;
r->x = w;
r->y = z;
retum(flag);
II IGPIB.CPP
#include <stdio.h>
#include <string.h>
#include <fcntl.h>
#include <io.h>
#include <sys\stat.h>
#include <ctype.h>

#include <stdlib.h>
#include '\cpp\k570\igpib\igpib.h"

 

 

1 53
struct g gpibError;
lr‘ GpibIO implantation ‘I
fpib_lO::Gpib_lO(void) II Constructor
char temp[81];

outfile = openCIgpibout",O_BlNARY|O_WRONLY);

if(outfile == -1){
sprintf(gpibError.errorMsg."Could not find gpibout");
gpibError.errorCode = -1;

infile = open("\gplbin".O_TEXT|O_RDONLY);

if(infile == -1){
sprintf(gpibErmr.errorMsg,'Could not find gpibin");
gpibError.errorCode = -2;

I:
if(ioctl(outfile,3,(void ‘)"BREAK",5) == -1)
I

strcpy(gpibError.errorMsg,_strerror("lOCTL error: '));
gpibError.errorCode = -3;

 

I;
Gpib_lO::~Gpib__lO(void) // Destructor
I
close(outfile); II Close device drivers.
close(infile);
I
W Gpib implantation I
//I/ Mode setting IIII

int Gpib::set_remote(void)

sprintf(str,"REMOTE %d”,device);
write _gpib(str);

retum(O);

I

int Gpib::set_auto(void)
{ .
sprintf(str,"OUTPUT %d;ROG1X",device);
wfiteribIstr);

retum(O);

 

154

int Gpib::set_local(void)

I
sprintf(str,"LOCAL %d",device);
write _gpib(str);

retum(O);

l/ll I/O l///

int Gpib::write_gpib(char *s)

I
char temp[81];
sprintf(temp,"%s\r\n",s);
if(wn‘te(outfile,temp,strlen(temp))== -1)

I
sprintf(gpibError.errorMsg,"Error writing to GPIB“);
gpibError.errorCode = -1;
retum(-1);

else return 0;
I
int Gpib::read_gpib(void) II Private
I
sprintf(str,"ENTER %d",device);
write _gpib(str);
if(read(infile,str,20) == -1)
sprintf(gpibError.errorMsg,”Error reading from GPIB");
gpibError.errorCode = -2;
retum(-2);
else

I
int i,j;
for(i=0; I<20;i++)
If(lscntrl(str[i])){ str[i] = 0; break;}
retum(O);

I:

I;

int Gpib::readeib(float8 x)
I

if(read_gpib0) retum(gpibErrorerrorCode);
x = atof(str);
retum(O);

int Gpib::readeib(int8 i)

 

155

 

I
if(read_gpib()) retum(gpibError.errorCode);
i = atoi(str); .
retum(O);
I:
int Gpib::readeib(char *s)
I
if(read_gpib()) retum(gpibError.errorCode);
strcpytsstr);
retum(O);
I:
I"
K570.CPP
This file contains the primitives of the HQ with the Kiethley
570 system.
*I
#include <dos.h>
#include <conio.h>
#include <stdio.h>
#include 'keithley.h"
i” ‘ “ -‘ ‘ ‘ --------- /
I‘
This constructor initializes the values in the structure K570
which describes the Keithley 570 hardware.
*I
K570::K570(int g, unsigned b)
I
I/ Locations
base = b;
select_slot = 0x01;
select_chn = 0x0A;
ad_low_data = 0x02;
ad_high_data = 0x03;
ad_status = 0x18;
ad _g _gain = 0x1A;
ad_slot = 0x06;
da_control = 0x04;

da_data = 0x05;

//

//

//

l/

//

//

I.

156

da_strobe = 0x1 D;
digi_in_a = 0x06;
digi_in_b = 0x07;
digi_out_a = 0x08;
digi_out_b = 0x09;
power_ctrl_a = 0x0C;
power_ctrl_b = 0x0D;
cmd_a_os = OXOE;
cmd_b_os = 0x0F;
Interface locations
rw_cnt_O = 0x40;
rw_cnt_I = 0x41;
rw_cnt_2 = 0x42;
cnt_control = 0x43;
timer _global = 0x60;
timer_status = 0x61;
clear_interrupt = 0x62;
set_int_level = 0x63;
Constants

ad_busy = 255;
ad_ready = 127;

Announce Initialized
printf("lnitialized\n');
Set the global gain

BYTE temp = 0;
gain = 9:

Analyze and set the gain

if(gain <= 1) I temp = 0: gain = 1;};

if((gain > 1) 88 (gain <5)) {temp = 1; gain = 2;}
“(team > 4) 8:8» (gain <10» { temp = 2; gain = 5;};
if(gain > 5) {temp = 3; gain = 10;};

pokeb(base,ad_g_gain,temp); II Set gain
pokeb(base,da_strobe,0x40); II Enable strobe

Low-level read AD converter. gets the channel number and
returns one of either two int values:

 

157
1) 0 to 2048 - if every thing was OK.
2) -255 to -1 - if user intenupt. This value is the
negative of the character pressed.
‘/
int K570::read_ad(int channel)
I
BYTE temp = 0;
unsigned result = 0;
II Select slot and channel
pokeb(base,select_slot,ad_slot);
pokeb(base,select_chn,(BYTE) channel);
for(result = 0;result<50;result++); II Delay
II printf("Selected slot and channel\n");
pokeb(base,ad_status,1); // Start a new conversion
dOI
I/ printf("Busy\n");
temp = peekb(base,ad_status); ‘ '
key = check_kbd0; II Check if key was pressed
if(key) retum(-key); II If so return the key
}while(temp I= ad_ready);
II printf("OK\n");
temp = peekb(base,ad_high_data); // Get data
result = temp << 8;
temp = peekb(base,ad_low_data);
result += (unsigned)temp;
result = result 8 0xFFF; II Mask 12 bit
pokeb(base,ad_status,1); I/ Start a new conversion
retum((int)result);
Boolean K570::write_da(int channel,int ivalue)

BYTE low = 0, high = 0;

II Break to two bytes

ivalue 8= 0xFFF; II Mask 12 bits

high = ivalue >> 8; II Upper 8 bits

low = ivalue 8 0xFF; II Lower 8 bits
II Load AD

switch(channel)

case 0: pokeb(base, da_control, 0); II Low chn 0

pokeb(base, da_data, low);

 

 

158

pokeb(base, da_control, 1); II High chn 0
pokeb(base, da_data, high);
break;

case 1: pokeb(base, da_control, 2); II Low chn 1
pokeb(base, da_data, low);
pokeb(base, da_control, 3); II High chn 1
pokeb(base, da_data, high);

break;
default : return false;
I:
pokeb(base,da_strobe,1); II issue data
return true;

I;

II Implantation of the MonitorUserIntr

int MonitorUserInt::check_kbd(void)
I

unsigned c=0;

if(kbhit0)c = getcho; II Check the KBD if has something get it.
retum(c);

.!;;AAAAAAAAAA-AAAAAA;--A---AAAAAAAAAAAA‘A‘----AAA;A-A;An-‘-AAAA‘AAAAA/

 

I
II Implentation of the functions in K570_Analog.
K570_Analog::K570_Analog(K570‘ k,
int ad_range, int da_range,
int ad_chn, Int da_chn)
I
daChannel = da_chn;
adChanneI = ad_chn;
hw = k; II Pointer to the K570 object.
II DA and DA transformation constants
switch(ad_range)
I
case 1: ad.mult = 0.002441F; II 0 - 10 V
ad.offset = 0;
break;
case 2: ad.mult = 0.001221 F; II 0 - 5 V
ad.offset = 0;
break;
case 3: ad.mult = 0.004882; II -10 - 10 V
ad .offset = -2048;
break;

case 4: ad.mult = 0.002441; l/ -5 - 5 V

 

 

159
ad.offset = -2048;
break;
case 5: ad.mult = 0.001221; II -2.5 - 2.5 V
ad.offset = -2048;
break;
default: ad.mult = 0.001221; /I -2.5 - 2.5 V
ad.offset = -2048;
I
switch(da_range)
case 1: da.mult = 409.67F; II 0 - 10 V
da.offset = 0;
break;
case 2: da.mult = 819.34F; II 0 - 5 V
da.offset = 0;
break;
case 3: da.mult = 204.83; II -10 - 10 V
da.offset = 2048;
break;
case 4: da.mult = 409.67; II -5 - 5 V
da.offset = 2048;
break;
case 5: da.mult = 819.34; II -2.5 - 2.5 V
da.offset = 2048;
break;
default: da.mult = 819.34; // -2.5 - 2.5 V
da.offset = 2048;
I
I
I* Read the actual voltage from the AD converter. The function returns
false if everything went OK or the character that was in the KBD.
Thus the call to this function is of the form:
user_interrupted = read_voltage(1,0.05F);
if(user_interrupted) request = user_internipted;
The value read is loaded as a reference variable ”value". If the
user interruped it is set to 0.0F.
*/
int K570_Analog::readVoltage( float8 value)
I
int temp;
value = 0.0F;
temp = hw->read_ad(adChannel);
if(temp < 0) retum(-temp); II User intr. Convert back to positive.
II Convert from word to float
value = (float)(temp + ad.offset);
value '= ad.mult;
retum(false);

I.

 

 

160

Boolean K570_Analog::writeVoltage( float8 value)
I
int v=0;
v = da.mult * value; II Convert to a word
v += da.offset;
if(lhw->write_da(daChannel,v)) l/ Write it
return false;
else
return true;

void K570_Analog::calibrate(float8 alpha, float8 beta)
I

float x1 ,x2; II Instrument voltage
float v,y,z; II Voltage read.

printf("Set value #1. Enter voltage : ”);

scanf("%t‘,8v);

writeVoltage(v);

printf("Set value #1. Please enter the instrument reading: ");
scanf("%f",8x1);

readVoltage(v);

printf("Set value #2. Enter voltage : ');

scanf("%f',8y);

writeVoltage(y);

printf("Set value #2. Please enter the instrument reading: ");
scanf(“%f“.8x2);

readVoltage(y);

printf(“Set to zero");
oetcho:
readVoltage(z);

alpha = 2;
beta = (x1- x2) I (v - y);

I/ printf("\nAIpha = %10.3G\nBeta = %10.3G\n",alpha,beta); II

I:
I‘

PRIMRITIV.CPP

Voltage and clamp primitives.
*/

#include'\cpp\k570\primitiv.h"

#include<math.h>

V0ltage::Voltage(K57O 'k,

int ar,
int dr,
int dc,
int ac,
float 5,
float m,
float 0,
float d
):K570_Analog(k,ar,dr,ac,dc)
I
dAV = d;
inMuIt = m;
inOffset = 0;
step = s;
cAV = 0.0F;
I:

II Change conversion factors.

161

void Voltage::changeCF(float m, float o)
inMult = m;
inOffset = o;
I;
int Voltage::getVoItage(float8 v)
I
float temp;
int i;
if(I(i = readVoltage(temp»)
v = (temp - inOffset) ' inMult; ~
else
v=00fi
retum(i);
I:
int Voltage::setVoItage(float v)
I
float cV;
int flag = false;
for(;;) II Forever.
I
flag = getVoltage(cV); II Read voltage (converted)

If(flag) retum(flag);
if(fabs(cV - v) <= dAV) retum(flag);
if(cV > v) cAV -= step;

II User interrupt
I/ Voltage is correct; exit.
II Change the applied real voltage.

 

162

 

else If(cV < v) cAV += step;
if(lwriteVoltage(cAV)) ./I Write real voltage.
retum(-1); // exit on enor in writing with -1.

I:
I:
I““““*‘*““""*“““““““‘*“**“““““‘*“““““.‘..‘..‘.““‘/
extem struct g gpibError;
Current::Current(int i, float m, float o):

Gpib(i)
I

mult = m;

offset = o;

set_remote();

set_autoo;
I:
void Current::ChangeCF(float m, float o)
I

mult = m;

offset = o;
I;
int Current::getCurrent(float8 i)
I

int flag;

flag = readeib(i);

if(flag) retum(gpibError.errorCode);

else {

i = i * mult + offset;
retum(false);
I;

I;

lClamp::lClamp(Current*i, Voltage *v, float di)
I
I = i;

V = v;
dCl = di;

int lClamp::clamp(float i,float 8v)
float clampl,cV,cl,R, clampV;
int flagi = false,flagv = false,sign;
I0"19 I':
clampl = i;

v->setVoltage(0.05);

1 63
for(i=0;j<1000000L;j++):

flagi = I->getCurrent(cl); II Read current
if(flagi) retum(flagi); II GPIB error

flagv = V->getVoltage(cV); II Read voltage
if(flagv) retum(flagv); II User interrupt

R = W I cl;

clampV = i * R;

(i > 0.0F) ? sign =1 : sign = -1:
if(fabs(clampV) > 0.190F) clampV = 0.199 ' float(sign);

for(;;) II Forever.
I
flagi = I->getCurrent(cl); II Read current
if(flagi) retum(flagi); II GPIB en’or
flagv = V->getVoltage(cV); II Read voltage
if(flagv) retum(flagv); // User inten'upt
if(fabs(cl - clampl) <= dCl)
I
v = cV;
retum(flagi); ll Current is correct; exit.

If(cl > clampl) clampV -= V->step; II Change the applied real voltage.
else if(cl < clampl) clampV += V->step;

if(V->setVoltage(clampV)) II Write real voltage.
I
for(i=0;j<1000000L;j++);
retum(-1); I/ exit on error in writing with -1.
I;
I;
/"
Userio.cpp
*/
#include<bios.h>

#include"userio.h"
II If there is a key waiting get the extended code

int userKey(void)

{

Il‘lt ans;

if(bioskey(CHECK))
ans = bioskey(GET)8 Oxff;
else ans = 0;

retum(ans);

I.

164

I" If there is a key waiting get the extended code. The code goes to
the lower byte. The upper byte contains the status of the control
keys like shift,ctrl etc..

'/
int userXKey(void)
I
int ans;
if(bioskey(CHECK))
I
ans = bioskey(GET)8 Oxff;
ans += (bioskey(GETXTRA) 80xff) << 8;
I
else ans = 0;
retum(ans);
I:
/*
W_TYPE.CPP
Implantation of W_TYPES.H
*I

#include '\cpp\k570\gwind\w_types.h"
#include <string.h>

II Setting the text parameters;

int textSter(struct textsettingstype s)

I
settextstyle(s.font,s.direction,s.charsize);
if(graphresulto == grOk)
settextjustify(s.horiz,s.vert);
else
retum(-1);
retum(O);
I
TextButton::TextButton(
int I, int u,
int r, int b, II Viewport
char atfl. II Active text
char iat[], II Inactive text.
int afc. II Active foreground color
int iafc, II Inactive foreground color
int abc, II Active background color
int Iabc, II Inactive background color
int d, II Text directions.

int hj, I/ Text justification.

I

165

int vj,

float x, II Origin location x
float y, II Origin location y
int f, II Font type

int sz II Size.

Gwindow(l,u,r,b)

II Load the interanl values:

specs.horiz =hj;
specs.vert = vj;
textStyle(specs); ”Open the text
”Open the window as inactive
inactivateo;
I:
int TextButton::activate(void)
I
textStyle(specs);
colors.bg = activeBgColor;
drawWino;
setcolor(activngColor);
outtextxy(xToi(Iocation.x),yToj(Iocation.y),activeText);
retum(O);
I:
int TextButton::inactivate(void)
I
textStyle(specs);
colors.bg = inActiveBgColor;
drawWino;
setcolor(inActivngColor);
outtextxy(xToi(location.x),yToj(Iocation.y),inActiveText);
retum(O);

I;

activngColor = afc;
inActivngColor = Iafc;
activeBgColor = abc;
inActiveBgCoIor = iabc;

location.x = x;
Iocation.y = y;
strcpy(activeText,at);
strcpy(inActiveText,iat);
status = inactive;

speesjont = f;
specs.direction = d;
specs.charsize = $2;

166

void TextButton::changeActiveText(const char 50)

I
strcpy(activeText,s);

void TextButton::changelnActiveText(const char s[])

I
strcpy(inActiveText,s);

int TextButton::changeColors(int 3, int b, int c, int d)
I

activngColor = a;

inActivngColor = b;

activeBgColor = c;

inActiveBgCoIor = d;
retum(O);

PlainTextWinzzPlainTextVIfin

( int I. int u,
int r, int b, II Viewport
char at[], II Text
enum visibility see, II Show or hide.
int save, II No
int afc, II Foreground color
int abc, I/ Background color
int d, II Text directions.
int hj, II Text justification
int vj,
float x, II Origin location x
float y, II Origin location y
int f, II Font type
int 52, II Size.
int br, II No border
int brfg,
int brbg

):
Gwindow(l,u,r.b,see,save,abc,brbg,brfg,afc,1,br,1)
I

fgCoIor = afc;

location.x = x;
Iocation.y = y;
strepy(text.at);

specs.font = f;
specs.direction = d;
specs.charsize = 52;
specs.horiz =hj;

 

 

167
specs.vert = vj;

textStyle(specs); ”Open the text
putTextO; II Will output only if visible
I:
void PlainTextWin::putTexKvoid)
I
if(visible) ll Otherwise don't waste, time
I
textStyle(specs);
setNeto;
setcolor(fgColor);
outtextxy(xToi(location.x),yToj(location.y),text);
I;
I;

void PlainTextWin::changeText(const char s[])

amt/(texts):
if(visible)
I
clearWino;
putTexto;
I:
H or fll s
It
conflg.h
System configuration IIO
*/
#include<stdio.h>

Example of configuration file ( in the file there are no comments):

5 II The AD range
5 II The DA range
-0.1 II V multiplier

0.0 II V offset
-1.0 II I Multiplier
0.0 II I Offset

1.0 II C V0

0.0 II C V residual
0.0 II C Cs

1.iv II Default file name

168

60000 II Total step time (mSec)
40000 II Delay

1000 II Interval

10 II # Measurrnents

7 II # steps

-0.4 II Minimun V

0.4 II Maximum V

0.1 /I Max +I- voltage (Volts)

-0.7 /I Min and max of the X axis

-2e-9 II Min and max of the Y axis

 

int adR,daR;/I AD and DA ranges.

II Conversion factors for V,l and C
float vMuIt;
float vOffset;
float iMult;
float iOffset;
float cVO;
float chesduaI;
float cCs;

char fn[81];
char cfn[81];

II IV parameters
long ivT; II Total step time (in mSec)
long ivD; II Delay time
long ivl; II Interval between measurrnents
int ivNM; [I Number of measunnents
int ivNS; II Number of voltage steps
float ivMinV; I/ Minimum Voltage
float ivMaxV; // Maximum voltage
float ivMV; I/ Maximum applied voltage;
float minX,maxX.minY,maxY;

protected:
F ILE‘ f;

int nB, nF, hF, hB;
int col;

int min,max;

void hiLight(int i);
void loLight(int i);
void getVar(int i);

public:

169

Config(char s[] = "K570.cnf'):
~Config(void);

void review(void):

void saveConfig(char 50);
void readConfig(void);

G_WIND.H
Graphic window class definitions.
*I

#include <graphies.h>

enum VGA_COLORS {VGA_BLACK,VGA_BLUE,VGA_GREEN,VGA_CYAN,VGA_RED,
VGA_MAGENTA,VGA_BROWN,VGA_LIGHTGRAY,VGA_DARKGRAY,
VGA_LIGHTBLUE,VGA_LIGHTGREEN, VGA_LIGHTCYAN,
VGA_LIGI-ITRED,VGA_LIGHTMAGENTA,VGA_YELLOW,VGA_WHITE};

enum visibility {hidden = 0, shown};

II Video information.

struct Video{
int maxx;
int maxy;
int maxcolor,
I;

void gettheoMax(struct Video *v);

II Viewport information.

struct Vport{
int left, up, right, bottom;
int x,y;
I:

/I Colors.

struct Wcolors{
int bg; II Background.
int bbg; II Border background.
int bfg; II Border foreground.
int current; II Current foreground color.
I;

l/ Dimensions of the window for the user. The window is assigned
II coordinates in real numbers.

struct Wdimentions{
int directions; /I 0 is normal, 1'is minimum at the lower left.
float minx,miny; II Size of the window in real numbers
float maxx,maxy;

170

I:
II Real number coordinates.
struct Rcoord{
float x,y;
I:
I‘ The Gwindow can be operated in two modes:
1. Simple, no hiding capability. If 'saved' is set to no (0)
there is no hiding capability.
2. If 'savebg' is set to yes (1). the windows background is save
upon call to showWindowo.
The 'savebg' parameter overrides the 'see' parameter. If 'savebg' is set
to no a call to hideWindow will be ignored.
*/
class Gwindow
I
private:
void far 'bgimage; II Background of the window
void far 'storedimage; II The window itself when hidden.
float xratio, yratio; I/ Convertion const from real dimentions
II to pixels.
int saved; II Bg has been saved?
int xpixels,ypixels; II In the viewport.
int clip;
protected:
stnict Vport total; II The total area of the window;
struct Vport net; II The net (without border)
II area of the window;
void setNet(void); II Set viewport to net
void setTotal(void); II Set viewport to total
void drawWin(void);
int xToi(float x); II Convert float X to i in the net VP
int yToj(float y): II Convert float Y toj in the net VP
public:

struct Wdimentions dim; II Dimentions of the window
struct Wcolors colors; II Colors;
int borderType: II Border type code (0 = none,

II1 = single line, 2 = Double line).
int stored; II Is the window stored? (0/1).
int hideCapabIe; II Is capable of hiding? (0/1).
enum visibility visible; II Is the window visible.

Gwindow(int I, int u.
Int r. int D, II Viewport
enum visibility see = hidden, // Open shown or hidden.
intsavebg=0, IIO=no1=yes

 

Int cbg = VGA_BLUE,
int cbbg = VGA_BLUE,
Int cbfg = VGA_RED,

int ca = VGA_WHITE,

int clp = 1,

int border = 0,

Int dir = 1,

float xmin = 0.0F.
float ymin = 0.0F.
float xmax = 1.0F,
float ymax = 1.0F

I:
~Gwindow(void);

void openthdow(void);
void closeWindow(void);
void showWindow(void);
void hideWindow(void);

I/ Where is the cursor?

171

I/ Bg color.
/I Border bg color.

.II Border foreground color.

II Active forgraound color.
I/ On.

II No border.

II Direction flag 0 = normal 1 = real.
II Dimensions.

int whereIsCrs(int x, int y,struct Rcoord *r);

int isanin(int x, int y);

int isanorkArea(int x, int y);

II is it in the window?
II or in the working area?

int setDimensions(float xmin, float ymin, float xmax. float ymax);

int drawLine(float x1, float y1, float x2, float y2, int color);
int drawPoint(float x, float y, int color);

void clearWin(void);

// Global function definitions:

int initGraphicant ‘gdriver, int 'gmode, char pathfl);
int iniLinkedGraphicant *gdriver, int *gmode);

II
HWCLOCK.H
*/

#ifdef _cplusplus
extem 'C"{
#endif

I'” Function declrations “*I
void intenupt far new_int8(void);
void new_speed(void);
void reset_speed(void);
void start_clock(void);
void reset_clock(void);

#ifdef _cplusplus

 

 

172

I
#endif
It
gpib.h
Class definitions to GPIB comunication.
*I
struct g II Error message structure
I

char errorMsg[81]:
int errorCode;

}.
It
The base class contains the handles to the device driver for
input and output. The constructor initializes the comunication
by an IO_CTRL.
‘7
class Gpib_IO
I
public:
int infile;
int outfile;
Gpib_lO(void); I/ Constructor
~Gpib_lO(void); I/ Destructor
Ii
Actions on the device. Set the single device to remote and local
and read and writes to it a string.
*I
class Gpib:Gpib_IO
I
int read _gpib(void);
public:
Int device; II The address of the device
char str[80]; /I String for communication

Gpib(int i = 22):Gpib_l00{device = i;};

int set_remote(void);
int set_local(void);
int set_auto(void);

int write _gpib(char *8);

Int readeib(float8 x);
int readeibGnt8 i);
int readeib(char ‘5);

If

1 73
IV.H */

#include'gwind\w_types.h'
#include"timer\hwclock.h"
#include"primitiv.h"
typedef float point[2];

class

{

public:

//
//
//
//

IV

II Graph window /I
Gwindow *graph;

II Text windows II
PlainTextWin *imax;
PlainTextWin 'imin;
PlainTextWin *ititle;
PlainTextWin *vmin;
PlainTextWin 'vmax;
PlainTextWin *vtitle;
PlainTextWin *nSteps;
PlainTextWin *flleName;

I'***"' Buttons MI

I“ AAAAAAAAAAAAAAAAAAAAAAAA /
TextButton *nrnswitch;
TextButton *status;

I**"'"*‘ Buttons WI

“A AAAA LAAAA-AAA LAAAAAAA/

l

PlainTextWin ‘msg;

 

float x,y,xo,y0;

float dV;

int step;

point "lable;

point “temp;

void measurment(void);

void drawCross(float x, float y, int color);
void adjustVoltage(void);

friend Voltage;
friend Current;
Voltage ‘V;
Current *I;

long stepTime,delayTime,intervaI;
Int nMeasurments, nOfSteps;
float minX,maxX,minY,maxY;
float MinV,MaxV,MinI;

char *fn;

void makeScreen(void);

void runTask(void);

int popUp(void);

keithley.h
This file contains the defineitions for l/O with the Kiethley
570 system.

 

 

174

#ifndef BOOLEAN

enum Boolean {false,true};
#define BOOLEAN

#endif

#include "userintr.h"
typedef unsigned char BYTE;

// Transformation constants
stnict Transform{

int offset;
float mult;

I:

static struct K570:MonitorUserlnt

I
unsigned base; II Base segment

II A to D
BYTE select_slot; II Select slot register
BYTE select_chn; II Select channel register
BYTE ad_low_data; II AID LSB register
BYTE ad_high_data; II AID MSB register
BYTE ad_status; II AID ststus register
BYTE ad 9 gain; II AID global gain register
BYTE ad_slot; I/ AID slot number

I/ D to A

BYTE da_control; II DIA control register
BYTE da_data; II DIA data register
BYTE da_strobe; II DIA strobe mode register

II Digital IIO ports
BYTE digi_in_a; II Digital input A register
BYTE digi_in_b; II Digital input B register
BYTE digi_out_a; II Digital output A register
BYTE digi_out_b; II Digital output B register
I/ Power control ports

BYTE power_ctrl_a; II Power control A register
BYTE power_ctrl_b; II Power control B register

BYTE cmd_a_os; I/ Option slot command A register
BYTE cmd_b_os; II Option slot command B register

I/ Interface locations

 

 

/I

175

BYTE rw_cnt_o; II ReadNVrite counter 0 register
BYTE rw_cnt_1; II Read/Write counter 1 register
BYTE rw_cnt_2; I/ ReadNVrite counter 2 register
BYTE cnt_control; ll Counter control register

BYTE timer _global; II Timer global configuration register
BYTE timer_status; II Timer status register

BYTE clear_interrupt; II Clear interrupt (570 system)
BYTE set_int_level; II Set intemrpt level register

A to D Constants

BYTE ad_busy; II AID is busy

BYTE ad_ready; II AID is done and ready for next
int gain; II Global AID gain value.
K570(int g = 1, unsigned b = 0xCFF8); .

int read_ad(int channel);

Boolean write_da(int channel,int ivalue);

class K570_Analog

I
II

//

public:

This class provide the means to read and write voltage values
friend K570; II Allow to use K570 variables
int adChanneI; // This task's analog input channel

int daChannel; II This task's analog output channel
K570‘ hw;

Transformation Constants

Transform ad; II Contains the AID transformation
II constnats. (From 12 bit to volts)
Transform da lI Contains the DIA tronsformation

II constnats. (From volts to 12 bit)

K570_Analog(K570 ‘k, /l Pointer to K570 object
int ad_range = 5, int da_range = 5, I/ Ranges
int da_chn = 0. int ad_chn = 0); II Channels

int readVoltage(float8 value);

Boolean writeVoltage(float8 value);

void calibrate(float8 alpha, float8 beta);

 

 

176

V“ LOGO.H ’/

#include <conio.h>
#include <string.h>

class Logo

Logo(void)
I

int i;
char s[81];

textbackground(BLUE):
textcolor(WHlTE);
clrscro;

strcpy(s,”BLM Electrical measurment system");
i = 40 - stnen(s) I2;

gotoxy(l, 1 0);

cput5(5):

strcpy(s,"Develomd by M. Zviman");
I = 40 - strlen(s) I2;

gotoxy(i, 12);

0906(5):

strcpy(s,"Dept. of Physiology");
i = 40 - strIen(s)12;

gotoxy(i, 1 3);

6911156):

strcpy(s,"Michigan State University");
i = 40 - strien(s) I2;

gotoxy(i, 14);

cont-9(5):

strcpy(s,"Compuserve ID# 740072232");
i = 40 - strien(s) I2;

gotoxy(l, 1 5);

OWNS):

strcpy(s,"Hit any key to continue....");
i = 40 - strlen(s) I2;

gotoxy(i, 23);

cpmS(S);

Getcho;
I:

 

 

177

 

I:
It
PRIMITIV.H
Contains primitives of voltage handaling, clamp etc.
'I
#include'\cpp\k570\keithley.h"
#include'\cpp\k570\igpib\igpib.h'
It

This class define handling of voltage of one channel. It gets one analog input and one analog
output channel. The member functions read the voltage from the input (converted voltage) and
provided that the input is a voltmeter can apply voltage. As a simple input the setVoltage
function should be avoided.

 

‘I
class Voltagezpublic K570_Analog
I
float dAV; II Voltage resulution.
float inMult;
float inOffset;
public:
float step; II Applied voltage step.
float cAV; I/ Current applied voltage
Voltage(K570* k,
int ar=5, II Defualted in the definition of
int dr=5, II the K570_analog class
int dc=0,
int ac=0,
float s=0.0015, II Applied voltage step.
float m=-0.1, // Multiplier
float o=0.0, II Offset
float d=0.00015 II Voltage accuracy (1 mV)
I: '
int setVoltage(float v); /I Set the voltage to a converted v
int getVoltage(float8 v); II Get converted voltage
void changeCF(float m, float 0); II Change the conversion factors.
I;
I" “““““““““““““““““““““““““““““““““““““““““““““““““ /
class Current:public Gpib
I
float mult;
float offset;
public:
Current(int i = 22, II GPIB device address.
float m = -1.0, II Multiplier
float o = 0.0 I/ Offset

I:

 

 

178

void Current::ChangeCF(float m, float 0);
int getCurrent(float8 i);

I:
class ICIamp
I
friend Current;
friend Voltage;
Current *I;
Voltage "V;
public:
float dCI: II Current accuracy
lCIamp(Current *i, Voltage *v, float di = 2e-13);
int clamp(float i,float8 v);
I:
I“ Timer.h *I
class Timer
I
volatile long S_CLOCKS[4]; II Global counters
void interrupt far (‘dos_int8)(...); I/ Old int8 handler
void interrupt far ('newlnt8)(...);
int reset_int8; I/ 11 interrupts before a system tick
int int_count; II Counter for the interrupts
void new_speed(void);
void reset_speed(void);
public:
Timer(void);
void startClocks(void);
void stopClocks(void);
I:
II USERINTR.H

//

Monitor KBD for user interruption of a process.

static class MonitorUserInt

I
public:

int key; II The key scan code that was pressed
int check_kbd(void);

W_types.h

 

Graphic window types.

*I

#include ”\cpp\k570\gwind\g_wind.h"

enum activity {inactive = 0, active};

TextButtonszlndow

struct textsettingstype specs;

activngColor“,
inActivngColor,
activeBgColor“,
inActiveBgColor,

Rcoord location;
activeText[51];
inActiveText[51];
status;

TextButton(int l, int u,

class

I

private:

public:
int
int
int
int
stmct
char
char
int
int
int
void
void
int

};

class

I

private:

int r, int b,

char atfl,

char iat[],

int afc = VGA_WHITE,
int iafc = VGA_BLACK,
int abc = VGA_RED,

int iabc = VGA_GREEN,
int d = HORlZ_DlR,

int hj = CENTER_TEXT,
int vj CENTER_TEXT,
float x = 0.5F,

float y = 0.5F,

int f = DEFAULT_FONT,

int sz=1
);

activate(void);
inactivate(void);

179

// Viewport
// Active text

l/ Inactive text.
// Active foreground color

'l/ Inactive foreground color

// Active background color
ll Inactive background color
I/ Text directions.

// Text justification

I/ Origin location x
// Origin location y
// Font type

// Size.

changeActiveText(const char sfl);
changelnAcliveText(const char sfl);
changeColors(int a, int b, int c, int d);

PlainTextWin: public Gwindow

 

 

public:

int

1 80
struct textsettingstype specs;

int fgColor“,

struct Rcoord location;
char text[51];

PlainTextVlfrn(int l, int u,

int r, int b, II Viewport

char at[], // Text

enum visibility see = hidden, I/ Show or hide.
int save = 0, I/ No

int afc = VGA_WHITE, // Foreground color

int abc = VGA_BLUE. II Background color

int d = HORIZ_DlR. ll Text directions.

int hj = CENTER_TEXT, /l Text justification
int vj = CENTER_TEXT,

float x = 0.5F, ‘II Origin location x
float y = 0.5F. ll Origin location y
int f= DEFAULT_FONT, ll Font type

int $2 = 1, ll Size.

int br = 0, II No border

int brfg = VGA_RED,
int brbg = VGA_BLUE
);

void changeText(const char 50);

int changeColors(int a, int b);
void putText(void);

textStyle(struct textsettingstype s);

 

 

REFERENCES

 

REFERENCES

Ambros—lngerson, J., J. R. Granger, G. Lynch.1990. Simulation of
palecortex performs hierarchical clustering. Science. 247: 1 344-1 348.

Amoore, J. A. 1982. Odor theory and odor classification in Fragrance
Chemistry. Theimer, E. T. editor. Academic Press. New York. 28-76.

Amoore, J. E., and D. Venstrom. 1967. Correlation between
stereochemical assessments and organoleptic analysis of odorous

compounds in Olfaction and Taste. Hayashi, T. editor. Pergamon Press,
Oxford. 3-17.

Anholt, R. H. 1988. Incorporation of the olfactory adenylate cyclase in
lyposomes. AChemS X (Abstracts), Chemical Senses. 13 no. 4:671.

Anholt, R. R. H. 1989. Molecular physiology of olfaction. American
Journal of Physiology. 257:01043-C1054.

Anholt, R. R. H., R. W. Farmer, C. A. Karavanich. 1989. Excitation by
odorants of olfactory receptor cells in Chemical Senses. Brand, J. G., J.
H. Teeter, R. H. Cagan, M. R. Kare editors. Marcel Dekker. New York.
347-361.

Apfelbach, R., D. Russ, B. M. Slotnick. 1991. Ontogenetic changes in
odor sensitivity, olfaction receptor area and olfactory receptor density in
the rat. Chemical Senses. 16 no. 3: 209- 218.

Bakalyar, H. A. and R. R. Reed. 1990. Identification of a specialized
adenylyl cyclase that may mediate odorant detection. Science. 250:1403-
1406.

Baker, H., M. Grillo, F. L. Margolis. 1988. Olfactory marker protein (OMP)
expression in hamster CNS: biochemical and immunocytochemical
characterization. AChemSX (Abstracts), Chemical Senses. 13 no. 4:673.

181

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

182

Beets, M. G. J. 1978. Odor and stimultant structure in Handbook of
Perception, Vol. VIA: Testing and Smelling. Carterette, E. C. and M. P.
Friedman, editors. Academic Press. New York.

Beets, M. G. J. 1982. Odor and stimultant structure in Fragrance
Chemistry: The science of the sense of smell. Theimer, E. T. editor.
Academic Press. New York. 77-122.

Boyle, A. G., Y. S. Park, T. Huque, R. C. Bruch, 1987. Properties of
phospholipase C in isolated olfactory cilia from the channel catfish
(lctalurus punctatus). Compertr’ve Biochemistry and Physiology. 88B No.
3:767-775.

Breer, H. and I. Boekhofi. 1991. Odorants of the same odor class activate
different second messenger patways. Chemical Senses. 16 no.1:19—29.

Breer, H., l. Boekhoff, E. Tareilus. 1990. Rapid kinetics of second
messenger formation in olfactory transduction. Nature. 345:65-68.

Brennan, P., H. Kaba, E. B. Keveme. 1990. Olfactory recognition: a
simple memory system. Science. 250:1223-1226.

Bruch, R. C., J. H. Teeter. 1990. Cyclic AMP links amino acid
chemoreceptors to ion channels in olfactory cilia. Chemical Senses. 15
no.4: 419- 430.

Buck, L. and R. Axel. 1991, A novel multigene family may encode odorant
receptors: a molecular basis for odorant recognition. Cell. 65: 1 75-1 87.

Buck, L. and R. Axel. 1991. A novel multigene family may encode odorant
receptors. Journal of General Physiology. 98:3a.

Cain, W. S. 1978. History of research on smell in Handbook of
Perception, Vol. VIA: Testing and Smelling. Carterette, E. C. and M. P.
Friedman, editors. Academic Press. New York. 197-229.

Caprio, J. 1978. Olfaction and taste in the channel catfish: an
electrophysiological study of the responses to amino acids and
derivatives. Journal of Comparative Physiology A. 123:357-371.

Caprio, J. and R. P. Byrd. 1984. Electrophysiological evidence for acidic,
basic and neutral amino acid olfactory receptor sites in catfish. Journal of
General Physiology. 84:403-422.

Castellucci, V. F. 1985. The chemical senses: taste and smell in Principle
of neural science 2ed edition. Kandel, E. R. and Schwartz, J. H. editors,
Elsevier, New York. 409-425.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

183

Chastrette, M., A. Elmouaffek, P. Sauvegrain. 1988. A multidimentional
statistical study of similarity between 74 notes used in perfumery.
Chemical Senses. 13 no. 2:295-305.

Chen, Z. and D. Lancet. 1984. Membrane proteins unique to vertebrate
olfactory cilia: candidates for sensory receptor molecules. Proceedings of
the National Academy of Scienses USA. 81 :1859-1 863.

Cohen, F. S., 1986. Fusion of liposomes to planar bilayers, in Ion channel
reconstitution. Plenum Press, New York.

Davies, J. T. 1971. Olfactory Theories in Handbook of sensory
physiology. Vol IV: Chemical senses, Part 1: Olfaction. Beidler, L. M.
editor. Springer-Verlag, Berlin. 322-350.

De Lean, A., Munson, P. J., Rodbard, D. 1978. Simultaneous analysis of
families of sigmoidal curves: application to bioassay, radioligand assay
and physiological dose-response curves. American Journal of Physiology.
235(2):E97-E102.

Dhallan, R. S., K. Yau, K. A. Schrader, R. R. Reed. 1990. Primary
structure and functional expression of a cyclic nucleotide-acivated
channel from olfactory neurons. Nature. 347:184-187.

Dionne, V. E. 1989. Odor detection and discrimination: can isolated
olfactory receptor neurons smell? in Chemical Senses. Brand, J. G., J. H.
Teeter, R. H. Cagan, M. R. Kare editors. Marcel Dekker. New York. 415-
426.

Dionne, V. E. 1991. Odor transduction and discrimination by isolated
olfactory receptor neurons. Journal of general Physiology. 98:16a.

Discussion in the international symposium on receptor events and
transduction in taste and olfaction. 1989. Chemical Senses. Brand, J. G.,
J. H. Teeter, R. H. Cagan, M. R. Kare editors. Marcel Dekker. New York.
406-407.

Dixon, W. J. and Massey, F. J. 1951. Introduction to statistical analysis.
McGraw-Hill Inc. New York.

Edwards, P. A. and P. C. Jurs. 1989. Correlation of odor intensities with
structural properties of odorants. Chemical Senses. 14 no. 2:281-291.

Enomoto, S., M. Kashiwayanagi, K. Kurihara. 1991. Liposomes having
high sensitivity to odorants. Biochemica et Biophysica Acta. 106227-12.

35.

36

37.
38.

39.

40.

41.

42.

43.

44.

45.

46.

184

Epstein, M. and Racker, E. 1978. Reconstitution of carbamylcholine-
dependent sodium ion flux and desensitization of the acetylcholine
dependent receptor from Torpedo californica. Journal of Biological
Chemistry. 253:6660-6662.

Fazzalari, F. A. 1978. Compilation of Odor and Taste Threshold Values
Data. American Society for Testing and Materials. Philadelphia.

F inney, D. J. 1976. Radioligand assay. Biometrics. 32:721-740.

Firestein, S. 1992. Physiology of transduction in the single olfactory
sensory neuron in Sensory Transduction. Corey, D., P. and Roper, S., D.
editors. The Rockefeller University Press. 61-71.

F irestein, S. and F. Werblin. 1989. Odor-induced membrane currents in
vertebrate olfactory receptor neurons. Science. 244279-82.

Firestein, S. and G. M. Shepherd. 1991. The role of cyclic AMP as a
second messenger in vertebrate olfactory transduction. AChemS XII
(Abstracts), Chemical Senses. 15 no. 5:574.

Freeman, W. J. 1987. Simulation of chaotic EEG patterns with dynamic
model of the olfactory system. Biological Cybernatics. 56: 1 39-1 50.

F rings, 8. and B. Lindemann. 1991. Current recordings from sensory cilia
of olfactory receptor cells in situ, l. The neuronal response to cyclic
nucleotides. Journal of General Physiology. 97:1 -16.

Frings, S. and B. Lindermann. 1991. Properties of cyclic nucleotide-gated
channels mediating olfactory transduction: sidness of voltage-dependent
blockage by Ca2+ ions, amiloride, D600, and diltiazem. Journal of
General Physiology. 98: 1 7a.

Frings, 8., S. Benz, 8. Lindemann. 1991. Current recordings from sensory
cilia of olfactory receptor cells in situ, II. The neuronal response to cyclic
nucleotides. Journal of General Physiology. 97:725-747.

Gesteland, R. C. 1978. The neural code: integrative neural mechanisms
in Handbook of Perception, Vol. VIA: Testing and Smelling. Carterette, E.
C. and M. P. Friedman, editors. Academic Press. New York. 197-229.

Gesteland, R. C., Kleene, S. J., Pun, R. 1991. Patch recording implication
for olfactory transduction. Chemical Senses. 15 no.5: 581.

47.

48.

49.

50.

51.

52.

53.

54.

55.

57.

58.

59.

185

Getchell, M. L., B. Zielinski, T. V. Getchell. 1988. Odorant and autonomic
regulation of secretion in the olfactory mucosa in Molecular Neurobiology
of the Olfactory System: molecular, membranous and cytological studies.
Margolis, F. L. and T. V. Getchelleditors” Plenum Press, New York. 71-
98.

Getchell, T. V. and M. L. Getchell. 1982. Physiology of vertabrate
olfactory chemoreception in Fragrance chemistry: the science of the
sense of smell. Theimer, E. T. editor., Academic Press, New York.

Getchell, T. V. and M. L. Getchell. 1990. Regulatory factors in the
vertebrate olfactory mucosa. Chemical Senses. 15 no. 22223-231.

Gluter, E. 1983. Orgenic Chemistry. Hebrew University Press. Jerusalem.
(in Hebrew)

Goulding, E., J. Ngai, A. Chess, R. Kramer, S. Colicos, R. Axel, S.
Siegelbaum. 1991 . Biophysical Journal. 59:391 a.

Guardabasso, V., Rodbard, D., Munson, P. J. 1987. A model-free
approach to estimation of relative potency in dose-response curve
analysis. American Journal of Physiology. 152:E357-E364.

Halasz, N. 1990. The vertebrate olfactory system: Chemical
neuroanatomy, function and development. Akadémiai Kiado, Budapest.

Hanke, W., 1986. Incorporation of ion channels by fusion, in Ion channel
reconstitution. Plenum Press, New York.

Hays, W. L. 1973.8tatistics for the social sciences. Holt, Rinehart and
Winston Inc. NewYork.

Huxtable, R. J. 1986. Biochemistry of Sulfur. Plenum Press. New York.

Jones, D. T. and R. R. Reed. 1989. Golf: An olfactory neuron specific-G
protein involved in odorant signal transduction. Science. 244:790-795.

Jones, D. T., 1990. Distribution of the stimulatory GTP-binding proteins,
Gs and Golf, within olfactory neuroepithelium. Chemical Senses. 15 No.
3:333-340.

Joshi, H., M. L. Getchell, B. Zielinski, T. V. Getchell. 1987.
Spectrophotometric determination of cation concentrations in olfactory
mucus. Neuroscience Letters. 82:321-326.

60.

61.

62.

63.

65.
66.

67.

68.

69.

70.

71.

72.

186

Kang, J. and Caprio, J. 1991. Electro-olfactogram and multiunit olfactory
receptor responses to complex mistures of amino acids in the channel
catfish, Ictalurus punctatus. Journal of General Physiology. 98:699-721.

Kaupp, U. B., W. Altenhofen, W. Bonigk, E. Eismann, W. Kraus, J.
Ludwig. 1991. Family of cyclic nucleotide-gated ionic channels. Journal of
general Physiology. 9815a.

Kleene, S. J., R. C. MacDonald, M. S. Lidow, R. C. Gesteland. 1986.
Memebrane channels of olfactory cilia. AChemS Vll (abstract) Chemical
Senses. 10:393.

Labarca,P., S. A. Simon, R. R. H. Anholt. 1988. Activation by odorants of
a multistate cation channel from olifactory cilia. Proceedings of the
National Academy of Sciences USA. 85:944-947.

Laffort, L., Etcheto, M., Patte, F., Marfaing, P. 1989. Implications of power
law exponent in synergy and inhibition of olfactory mixtures. Chemical
Senses. 14 no. 1: 11-23.

Lancet, D. 1984. Molecular view of olfactory reception. TINS. 35-36.

Lancet, D. 1986. Vertebrate olfactory reception. Annual Reviews of
Neuroscience. 92329-355.

Lancet, D. 1988. Molecular components of olfactory reception and
trasduction in Molecular Neurobiology of the Olfactory System: molecular,
membranous and cytological studies. Margolis, F. L. and T. V. Getchell,
editors, Plenum Press, New York. 25-50.

Lancet, D. and U. Pace. 1987. The molecular basis of odor recognition.
TIBS. 12:63-66. ‘

Lerner, M. R., J. Reagan, T. Gyorgyi, A. Roby. 1988. Olfaction by
melanophores: what does it mean? Prooedings of the National Academy
of Sciences USA. 85:261-264.

Leveteau, J. l. Andriason, D. Trotier, P. MacLeod. 1989. Role of divalent
cations in EOG generation. Chemical Senses. 14 no. 5:611-620.

Levy, N. S., Bakalyar, H. A., R. R. Reed. 1991. Moleculae components of
the olfactory signal transduction pathway. Journal of General Physiology.
98:3a.

Linck, R. W. 1973. Comparitive isolation of cilia and flagella from the
lamellibranch mollusc Aequipecten irradians. Journal of Cell Science.
12:345-367.

73.

74.

75.

76.

77.

78.

79.

80.

81

82.

83.

187

Lowe, G. and G. H. Gold. 1991. The spatial distribution of olfactory
receptor currents. Chemical Senses. 15 no. 5:658.

Lynch, J. W. and P. H. Barry. 1989. Action potentials initiated by single
channels oppening in a small neuron (rat olfactory receptor). Biophysical
Journal. 55:755-768.

Lynch, J. W. and P. H. Barry. 1991. Properties of transient K+ currents
and underlying single potasium channels in rat olfactory receptor
neurons. Journal of General Physiology. 97:1043-1072.

MacKay, A. and P. Shaman. 1984. Topographical coding of odorant
quality is maintained at different concentrations in the salamander
olfactory epithelium. Brain Research. 297:207-216.

Mahan, B. H. 1975. University Chemistry 3rd edition. Addison-Wesley,
Reading.

Margolis, F. L. and T. V. Getchell. 1991. Receptor: Current Status and
Future Directions in Perfumes: Art, Science and Technology. Muller, P. M.
and D. Lamparski editors. Elsevier Applied Science. New York. 481 -498.

Margolis, F. L., 1988. Molecular cloning of olfactory-specific gene
products in Molecular Neurobiology of the Olfactory System: molecular,
membranous and cytological studies. Margolis, F. L. and T. V. Getchell
editors, Plenum Press, New York. 237-268.

Maue R. A. and V. E. Dionne. 1988. Membrane properties of isolated
olfactory receptor neurons in Molecular neurobiology of the olfactory
system. Margolis F. L. and T. V. Getchell eds. Plenum Press, New York.

Menevse A., G. Dodd, M. Poynder. 1978. A chemical-modification
approach to the olfactory code. Studies with a thiol-specific reagent.
Biochemical Journal. 176: 845- 854.

Miller, C. and Rosenberg, R. L., 1979. Modification of a voltage-gated K+
channel from sarcoplasmic reticulum. Journal of General Physiology.
74:457-478.

Miyamoto, T., D. Restrepo, J. H. Teeter. 1992. Voltage-dependent and
odorant-regulated currents in isolated olfactory receptor neurons of the
channel catfish. Journal of general Physiology. 99:505-530.

Munson, P. J., Rodbard, D. 1980. LIGAND: a versatile computerized
approach to characterization of ligand-binding systems. Analytical
Biochemistary. 107:220-239.

 

85.

87.

88.

89.

90.
91.

92.

93.

95.

188

Nakamura, T. and G. H. Gold. 1987. A cyclic nucleotide-gated
conductance in olfactory receptor cilia. Nature. 325:442-444.

Nelson, N., Lindstrom, J., Montal, M. 1980. Reconstitution of purified
acetylcholine receptors with functional channels in planar lipid bilayers.
Proceedings of the National Academy of Sciences U. SA. 77 no 5:3057-
3061.

Niles, W. D. and F. S. Cohen. 1987. Vidoe fluorescence microscopy
studies of phospholipid vesicle fusion with a planar phospholipid
membrane. Journal of General Physiology. 90:703-735.

Nomura, T. and K. Kurihara. 1988. Similarity of ion dependence of
responses to odorants between lipid bilayer membranes and olfactory
cells. JASTS >00 (abstract) Chemical Senses. 13 no. 3:315.

Ohloff G. 1982, The fragrance of ambergris in Fragrance chemistry: the
science of the sense of smell. Theimer, E. T. ed., Academic Press, New
York.

Ohloff, G. 1986. Chemistry of odor stimuli. Experientia. 42:271-279.

Ohloff, G., B. Winter, C. Fehr. 1991. Chemical Classification and
Structure-Odour Relationship in Perfumes: Art, Science and Technology.

Muller, P. M. and D. Lamparski editors. Elsevier Science Publishers. New
York. 287-330.

Pace, U. E. Hanski, Y. Salomon, D. Lancet. 1985. Odorant-sensitve
adenylate cyclase may mediate olfactory reception. Nature. 316:255-258.

Pelosi P, R. Maida. 1990. Odorant-Binding Proteins in Vertabrate and
insects: Similarities and Possible Common Function. Chemical Senses.
15 no. 2: 205- 215.

Pelosi, P. and R. Tirindelli. 1989. Structure/activity studies and
characterization of an odorant-binding protein in Chemical Senses.
Brand, J. G., J. H. Teeter, R. H. Cagan, M. R. Kare editors. Marcel
Dekker. New York. 207-226.

Pevsner, J. and S. H. Snyder. 1990. Odorant-binding protein: odorant
transport function in the vertebrate nasal epithelium. Chemical Senses. 15
no. 2:217-222.

Polak, E. H. 1973. Multiple profile-multiple receptor site model for
vertebrate olfaction. Journal of Theoretical Biology. 40:469-484.

 

 

97.

98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

189

Press, W. H., Flannery, B. P., Teukolsky, S. A., Vetterling, W. T., 1988.
Numerical recipes in C: The art of scientific computing. Cambridge
University Press. Cambridge.

Price, J. L. 1987. Thr central olfactory and accessory olfactory systems in
Neurobiology of taste and smell. Finger, E. T. and Silver, W. L. eds, John
Wiley & Sons, New York.

Restrepo, D. and Teeter, J. H., 1990. Olfactory neurons exhibit
hetrogeneity in depolarization-induced calcium changes. America! Journal
of Physiology. 258:C1051-C1061.

Restrepo, D. and Teeter, J. H., 1991. Initial characterization of the role of
inisitol-1,4,5-triphosphate in olfactory transduction in isolated rat olfactory
neurons. Journal of general Physiology. 98:173.

Restrepo, D., T. Miyamoto, B. P. Bryant, J. H. Teeter. Odor stimuli trigger
influx of calcium into olfactory neurons in the channel catfish. 1990.
Science. 249:1166-1 168.

Restrepo, D., Teeter, J. H., Honda, E., Boyle, A. G., Marecek, J. F.,
Prestwich, D., Kalinolski, D. L., 1992. Evidence for lnsP3 gated channel
protein in isolated rat olfactory cilia. American Journal of Physiology.
263:C667-C673.

Rhein L. D., R. H. Cagan. 1980. Bichemical Studeis of Olfaction: Isolation,
Characterization, and Odorant Binding Activity of Cilia from Rainbow
Trout Olfactory Rosettes. Proceedings of the National Academy of
Science USA. 77 No.8: 4412- 4416.

Rogers, K. E., P. Dasgupta, U. Gubler, M. Grillo, Y. S. Knew-Goodall, F.
L. Margolis, 1987. Molecular cloning and sequencing of a cDNA for
olfactory marker protein. Proceedings of the National Academy of
Sciences USA. 84:1704-1708.

Rosenberg, B., T. N. Misra, R. SWitzer. 1968. Mechanism of olfactory
transduction. Nature. 217:423-427.

Rosneberg, R. L., Hess, P., Reevs, J. P., Smilowitz, M., Tsien, R. W.,
1986. Calcium channels in planar lipid bilayers: insights into mechanisms
of ion permeatiom and gating. Science. 231:1564-1566.

Schild, D. 1988. Principles of odor coding and a neural network for odor
discrimination. Biophysical Journal. 54:1001-101 1 .

 

 

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

190

Schild, D., J. A. DeSimone, S. Hellwing. 1989. Excitation and adaptation
of frog olfactory receptor neurons upon stimulation with second
messengrs and natural odorants in Chemosensory Information
Processing. Schild, D. editor. Springer-Verlag. Berlin. 9-20.

Schindler, H and Quast, U. 1980. Functional acetylcholine receptor from
Torpedo marmorata in planar membranes. Proceedings of the National
Academy of Sciences U. SA. 77 no. 523052-3056.

Scott, J. W. and Harrison T. A. 1987. The olfactory bulb: anatomy and
physiology in Neurobiology of taste and smell. Finger, E. T. and Silver, W.
L. eds, John Wiley & Sons, New York.

Shepherd, G. M. 1991. Information flow in the olfactory sensory neuron.
Jounal of General Physiology. 98:2a.

Shepherd, G. M. 1992. Toward a consensus working model for olfactory
transduction in Sensory Transduction. Corey, D., P. and Roper, S., D.
editors. The Rockefeller University Press. 19-37.

Shirley S., E. Polak, G. H. Dodd. 1983. Chemical-modification studies on
rat olfactory mucosa using a thiol-specific reagent and enzymatic
iodination. European Journal of biochemistary. 132: 485- 494.

Shirley, S. G., J. Robinson, K. Dickinson, R. Aujla, H. Dodd. 1986.
Olfactory adenylate cyclase of the rat. Biochemical Journal. 240:605-607.

Shoji, T., and K. Kurihara. 1991. Sensitivity and transduction
mechanisms of responses to general odorants in turtle vomeronasal
system. Journal of General Physiology. 98:909-919.

Silver, W. L. 1987. The common chemical sense in Neurobiology of taste
and smell. Finger, E. T. and Silver, W. L. eds, John Wiley & Sons, New
York.

Simon, S. M. and G. Blobel. 1991. A protein-conducting channel in the
endoplasmatic reticulum. Cell. 65:371-380.

Simon, S. M., G. Blobel. J. Zimmerberg. 1989, Large aqueous channels in
membrane vesicles derived from the rough endoplasmatic reticulum of
canine pancreas or the plasma membrane of Eschrichia coli. Proceedings
of the National Academy of Sciences USA. 86:6176-6180.

Sklar, P. B., R. R. H. Anholt, S. H; Snyder. 1986. The odorant-sensitive
adenylate cyclase of olfactory receptor cells: differential stimulation by
distinct classes of odorants. The Journal of Biological Chemistry. 261 No.
33: 1 5538-1 5543.

 

120.

121.

122

123.

124.

125.

126.

127.

128.

129.

130.

131.

191

Smith, J. S., Coronado, R., Meissner, G., 1985. Sarcoplasmic reticulum
contains adenin nucleotide-activated calcium channels. Nature, 3161446-
449.

Snyder, S. H., P. B. Sklar, J. Pevsner. 1988. Molecular mechanisms of
olfaction. The Journal of Biological Chemistry. 263 no. 28:13971-13974.

Suzuki, M. 1987. Voltage-Dependent conductances in solitary olfactory
receptor cells. Annals of the New York Academy of Science. 510: 647.

Suzuki, N. 1989. Voltage and cyclic nucloetide-gated currents in isolated
olfactory receptor cells in Chemical Senses. Brand, J. G., J. H. Teeter, R.
H. Cagan, M. R. Kare editors. Marcel Dekker. New York. 469-494.

Takagi, S. F. 1978. Biophysics of smell in Handbook of Perception, Vol.
VIA: Testing and Smelling. Carterette, E. C. and M. P. Friedman, editors.
Academic Press. New York. 233-243.

Tien, H. T., 1974. Bilayer lipd membranes (BLM): Theory and practice.
Marcel Dekker. New York.

Tortier, D. 1986a. A patch-clamp analysis of membrane currents in
salamander olfactory receptor cells. Pfluger Archiv. 407:589-595.

Tortier, D., J. F. Rosin, P. MacLeod. 1989. Channel activites in in vivo and
isolated olfactory receptor cells in Chemical Senses. Brand, J. G., J. H.
Teeter, R. H. Cagan, M. R. Kare editors. Marcel Dekker. New York. 427-
448.

Tortier. D. and P. MacLeod. 1986b. cAMP and cGMP open channels and
depolarize olfactory receptor cells. ISOT IX abstract Chemical Senses.
1 1:674. -

Trombley, P. O. and G. L. Westbrook. 1991. Voltage-gated currents in
identified rat olfactory receptor neurons. The Journal of Neuroscience.
1 1(2):435-444.

Vassilev, P. M. and Hume, J. R., 1986. Calcium channels from cardiac
sarcoplasmic reticulum incorporated into bilayer membranes. Journal of
General Physiology. 88:60a.

Vassilev, P. M., Kanazirska, M. P., Tien H. T., 1986a. Calcium channels
from intercellular membranes reconstituted in patch-clamped bilayers.
Biophysical Journal. 49: 349a.

 

 

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

142.

143.

192

Vassilev, P. M., Kanazirska, M. P2, Tien H. T., 1987. Ca2+ channels from
brain microsomal membranes reconstituted in patch-clamped bilayers.
Biochemica et Biophysica Acta. 897:324-330.

Vodaynoy, V. 1989. Cyclic nucleotide-gate electrical activity in olfactory
receptor cells in Chemical Senses. Brand, J. G., J. H. Teeter, R. H.
Cagan, M. R. Kare editors. Marcel Dekker. New York. 319- 346.

Vodyanoy V, I. Vodyanoy. 1986. Electrical properties of chemically-gated
cation channels from olfactory epithelium homogenates. Biophysical
Journal. 492521 a.

Vodyanoy V, I. Vodyanoy. 1987. ATP and GTP are essential for olfactory
response. Neuroscience Letters. 73: 253- 258.

Vodyanoy V, I. Vodyanoy. 1988. Non-random fluctuations and ATP-
dependent dwell time of cAMP-gated ion channels from olfactory receptor
functionally reconstituted into bilayers. Biophysical Journal. 53: 505a.

Vodyanoy, V. 1982. Solvent-free lipid blmolecular membranes of large
surface area. Biochemica et Biophysica Acta. 687: 1 89-1 94.

Vodyanoy, V. 1991. Cyclic AMP-selective ion channels in olfactory
receptor cells. Chemical Senses. 16 no. 22175-180.

Vodyanoy, V. and R. B. Murphy. 1983. Single-Channel fluctuations in
blmolecular lipid membranes induced by rat olfactory eoithelial
homogenates. Science. 220271 7-719.

Vogt, R. G., R. Rybczynski, and M. R. Lerner. 1989. The biochemistry of
odorant reception and transduction in NATO ASI series Vol. H 39:
Chemosensory Information Processing. D. Schild, editor. Springer-Verlag,
Berlin. 33-76.

Wilkins, T. A., Chadney, D. C., Bryant, J., Palmstrom, S. H., Winder, R. L.
1978. Non-linear least square curve-fitting of a simple theoretical model to
radioimmunoassay dose-response data using a mini-computer in
Radioimmunoassay and related procedures in medicine. Vol I,
International Atomic Energy Agency. Vienna. 399-420.

Winegar, B. D., E. R. Rosick, R. Schafer. 1988. Calcium and olfactory
transduction. Comparative Biochemistry and Physiology. 91 A no. 22309-
315.

Wysocki, C. J. and Merdith, M. 1987. The vomeronasal system in
Neurobiology of taste and smell. Finger, E. T. and Silver, W. L. editors,
John Wiley 8. Sons, New York. 125-150.