Cellular and Molecular Basis of Mitochondrial

Inheritance Mitochondrial Disease and Fitness

1st Edition by Peter Sutovsky 9783030045708
3030045706 download
https://ebookball.com/product/cellular-and-molecular-basis-of-
mitochondrial-inheritance-mitochondrial-disease-and-fitness-1st-
edition-by-peter-sutovsky-9783030045708-3030045706-20448/

Instantly Access and Download Textbook at https://ebookball.com

 Get Your Digital Files Instantly: PDF, ePub, MOBI and More
Quick Digital Downloads: PDF, ePub, MOBI and Other Formats

Robbins and Cotran Pathologic Basis of Disease 9th edition by Vinay

Kumar, Abul Abbas, Jon Aster 9780323314817 0323314813

https://ebookball.com/product/robbins-and-cotran-pathologic-
basis-of-disease-9th-edition-by-vinay-kumar-abul-abbas-jon-
aster-9780323314817-0323314813-17866/

Robbins and Cotran Pathologic Basis of Disease 10th Edition by Vinay

Kumar, Abul Abbas, Jon Aster 0323609945 9780323609944

https://ebookball.com/product/robbins-and-cotran-pathologic-
basis-of-disease-10th-edition-by-vinay-kumar-abul-abbas-jon-
aster-0323609945-9780323609944-25788/

Robbins and Cotran Pathologic Basis of Disease 10th Edition by Vinay

Kumar, Abul Abbas, Jon Aster ISBN 9780323609944 0323609945

https://ebookball.com/product/robbins-and-cotran-pathologic-
basis-of-disease-10th-edition-by-vinay-kumar-abul-abbas-jon-
aster-isbn-9780323609944-0323609945-224/

Robbins and Cotran Pathologic Basis of Disease Professional 8th

edition by Kumar Abbas Fausto Aster ISBN 1416031219 978-1416031215

https://ebookball.com/product/robbins-and-cotran-pathologic-
basis-of-disease-professional-8th-edition-by-kumar-abbas-fausto-
aster-isbn-1416031219-978-1416031215-7166/
Robbins and Cotran Pathologic Basis of Disease 10th edition by Vinay
Kumar, Abul Abbas, Jon Aster ISBN 032353113X 978-0323531139

https://ebookball.com/product/robbins-and-cotran-pathologic-
basis-of-disease-10th-edition-by-vinay-kumar-abul-abbas-jon-
aster-isbn-032353113x-978-0323531139-3752/

Biochemistry The Molecular Basis of Life 7th Edition by James McKee,

Trudy McKee ISBN 0190847719 9780190847715

https://ebookball.com/product/biochemistry-the-molecular-basis-
of-life-7th-edition-by-james-mckee-trudy-mckee-
isbn-0190847719-9780190847715-8604/

Pathology of Heart Disease in the Fetus Infant and Child Autopsy

Surgical and Molecular Pathology 1st edition by Michael Ashworth ISBN
1107116287 978-1107116283

https://ebookball.com/product/pathology-of-heart-disease-in-the-
fetus-infant-and-child-autopsy-surgical-and-molecular-
pathology-1st-edition-by-michael-ashworth-
isbn-1107116287-978-1107116283-3596/

Robbins Cotran Pathologic Basis of Disease 10th edition by Vinay

Kumar, Abul Abbas, Jon Aster 0323609945 9780323609944

https://ebookball.com/product/robbins-cotran-pathologic-basis-of-
disease-10th-edition-by-vinay-kumar-abul-abbas-jon-
aster-0323609945-9780323609944-25802/

From Molecules to Networks An Introduction to Cellular and Molecular

Neuroscience 1st Edition by Ruth Heidelberger, Neal Waxham, John
Byrne, James Roberts ISBN 9780080491356 0080491359

https://ebookball.com/product/from-molecules-to-networks-an-
introduction-to-cellular-and-molecular-neuroscience-1st-edition-
by-ruth-heidelberger-neal-waxham-john-byrne-james-roberts-
isbn-9780080491356-0080491359-11668/
Advances in Anatomy, Embryology and Cell Biology

Peter Sutovsky Editor

Cellular and
Molecular Basis
of Mitochondrial
Inheritance
Mitochondrial Disease and Fitness
Advances in Anatomy, Embryology and Cell Biology publishes critical reviews and state-of-
the-art surveys on all aspects of anatomy and of developmental, cellular and molecular
biology, with a special emphasis on biomedical and translational topics.

The series publishes volumes in two different formats:

• Contributed volumes, each collecting 5 to 15 focused reviews written by leading experts
• Single-authored or multi-authored monographs, providing a comprehensive overview of
their topic of research

Manuscripts should be addressed to

Editor-in-Chief

Prof. Dr. P. SUTOVSKY, S141 Animal Science Research Center, Division of Animal Sciences and Department of
Obstetrics, Gynecology and Women’s Health, University of Missouri, Columbia, MO, USA
e-mail: [email protected]

Series Editors

Prof. Dr. F. CLASCÁ, Department of Anatomy, Histology and Neurobiology

Universidad Autónoma de Madrid, Ave. Arzobispo Morcillo s/n, 28029 Madrid, Spain
e-mail: [email protected]

Prof. Dr. Z. KMIEC, Department of Histology and Immunology, Medical University of Gdansk,
Debinki 1, 80-211 Gdansk, Poland
e-mail: [email protected]

Prof. Dr. H.-W. KORF, Anatomy and Brain Research Center, Department for Anatomy 1, Heinrich Heine University
Düsseldorf, Universitätsstrasse 1, 40225 Düsseldorf, Germany
e-mail: [email protected]

Prof. Dr. M.J. SCHMEISSER, Institute of Microscopic Anatomy and Neurobiology, University Medical Center of the
Johannes Gutenberg University, Langenbeckstr 1, 55131 Mainz, Germany
e-mail: [email protected]

Prof. Dr. B. SINGH, Faculty of Veterinary Medicine, University of Calgary, Calgary, AB T2N 4Z6, Canada
e-mail: [email protected]

Prof. Dr. J.-P. TIMMERMANS, Laboratory of Cell Biology and Histology/Core Facility Biomedical Microscopic
Imaging, Department of Veterinary Sciences, University of Antwerp, Drie Eiken Campus, Universiteitsplein 1,
2610 Wilrijk, Belgium
e-mail: [email protected]
231
Advances in Anatomy,
Embryology
and Cell Biology

Editor-in-Chief
P. Sutovsky

Series Editors
F. Clascá • Z. Kmiec • H.-W. Korf •
M.J. Schmeisser • B. Singh • J.-P. Timmermans

More information about this series at

http://www.springer.com/series/102
Peter Sutovsky
Editor

Cellular and Molecular Basis

of Mitochondrial Inheritance
Mitochondrial Disease and Fitness
Editor
Peter Sutovsky
Division of Animal Sciences
and Department of Obstetrics,
Gynecology and Women’s Health
University of Missouri
Columbia, MO, USA

ISSN 0301-5556 ISSN 2192-7065 (electronic)

Advances in Anatomy, Embryology and Cell Biology
ISBN 978-3-030-04569-2 ISBN 978-3-030-04570-8 (eBook)
https://doi.org/10.1007/978-3-030-04570-8

© Springer Nature Switzerland AG 2019

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, express or implied, with respect to the material contained herein or for any errors
or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims
in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface: A Mother’s Gift

Mitochondria, the cellular power station organelles, arose from bacterial endosym-
bionts and, in the course of evolution, became the essential source of the energy
storing substrate ATP in both prokaryotic and eukaryotic cells. Contrary to equal
maternal and paternal contribution of chromosomes in sexually reproducing organ-
isms, most animal taxa show preference for clonal, uniparental, and overwhelmingly
maternal inheritance of mitochondria and mitochondrial DNA (mtDNA). Research
into maternal mitochondrial inheritance, referred to as the “mitochondrial Eve
paradigm,” dates back to a 1940 Nature paper that explained it as a dispersion and
dilution due to disproportionate number of mitochondria in an oocyte, compared to a
spermatozoon (reviewed in Song et al. 2016a; Sutovsky and Song 2017). More
recently, studies in rodents, ungulates, and primates revealed the proactive, seek-
and-destroy mechanisms that target paternal, sperm-borne mitochondria after fertil-
ization (Sutovsky et al. 1996) with the help of the universal protein recycling
machinery, the ubiquitin-proteasome system (UPS) (Sutovsky et al. 1999, 2003).
Finally, mitochondrial membrane ubiquitination has been linked to autophagic
machinery in invertebrate genetic models Drosophila and C. elegans (Al Rawi
et al. 2011, 2012; Politi et al. 2014), eventually revealing that the linkage of
ubiquitination to autophagy also regulates postfertilization sperm mitophagy in
mammals (Song et al. 2016b). Besides advancing our knowledge of the develop-
mental and evolutionary aspects of mitochondrial function, a deeper understanding
of mechanisms promoting clonal mitochondrial inheritance in humans and other
species will help manage human mitochondrial health, fitness, and disease, as well as
improve production traits and reproductive performance in agriculturally important
domestic animals. This volume addresses the diverse yet interwoven aspects of
mitochondrial inheritance and function in animals and humans. Readers will notice
that the table of contents loosely follows the phylogenetic tree, starting with nem-
atodes and ending with humans. However, the topics of individual chapters are
complementary rather than overlapping, addressing the mitophagy mechanism,
mitochondrial genome interactions with environment and nutrition, and the impor-
tance of mitochondrial function for health, fitness, and reproduction.

v
vi Preface: A Mother’s Gift

Following early work on the role of UPS in the degradation of sperm mitochon-
dria at fertilization, the studies in C. elegans (Al Rawi et al. 2011, 2012; Sato and
Sato 2011) elegantly (pun intended) linked UPS, which degrades protein molecules
one at a time, to autophagy, the self-engulfment machinery that recognizes ubiquitin-
tagged protein aggregates and even whole organelles, including mitochondria, for
bulk degradation in the autophagic vacuole. The chapter by Jorge Merlet and
coauthors from the laboratory of Vincent Galy (Merlet et al. 2019) uses this animal
model, advantageous for its relative ease of genome modification and rapid gener-
ational turnover, to decipher the pathways that control the degradation of paternally
contributed sperm organelles. Fertilization of C. elegans oocyte occurs inside the
transparent body, which is advantageous for epifluorescence imaging of molecules
involved in the fertilization process. The mature oocyte passes through the sperma-
theca and an amoeboid spermatozoon cell fuses with the oocyte and delivers, among
the structures entering the embryo, the sperm mitochondria surrounded by the Golgi-
derived, nematode-specific membranous organelles. Subsequently, these sperm-
contributed structures are degraded by autophagy, guided by autophagic marker
proteins LGG-1 and LGG-2, homologues of mammalian autophagy receptors
GABARAP and LC3, respectively. This process may be facilitated by depolarization
and proteasome-dependent disruption of mitochondrial membrane, as well as by
mtDNA-specific endonucleases. The existence of integrated but distinct mechanisms
that converge during sperm mitophagy is proposed, opening up new lines of
investigation using both genetic and proteomic approaches.
One of the most notable exceptions to maternal inheritance rule is the bivalvian
mollusk Mytilus, in which the doubly uniparental inheritance is controlled by
specific genetic and molecular mechanisms guiding postfertilization recognition of
paternal mitochondria, reviewed in the second chapter by Eleftherios Zourous and
George C. Rodakis (Zouros and Rodakis 2019). A Mytilus egg comes either with or
without the “masculinizing” factor and the “sperm mitochondria binding” factor, a
property of the female that is determined by two corresponding nuclear genes that
are always in the on/on or the off/off phase. A fertilized egg without these factors
develops into a female and inherits sperm mitochondria while still having predom-
inantly maternal mtDNA. The egg endowed with said factors develops into a male,
and the sperm mitochondria become segregated into one embryonic blastomere
destined to give rise to sperm progenitor germ cells. In this case, paternal mitochon-
dria in the male embryo escape elimination from the germ line, resulting in sperma-
tozoa free of maternal mtDNA, while the soma of the embryo still inherits
predominantly maternal oocyte mitochondria. This complex, intriguing inheritance
pattern, further complicated by the possibility of mtDNA recombination in mussels,
thus flips the insect, nematode, and mammalian models of mitophagy by eliminating
maternal mtDNA from male germ cells during embryo development instead of
eliminating paternal mitochondria from fertilized egg after fertilization. Further
studies of this model will help answer question about the (co)evolution of genes
responsible for sexual reproduction, sex determination, nuclear–mtDNA interac-
tions, and allocation of reproductive efforts between sexes.
Preface: A Mother’s Gift vii

Continuing with the biparental inheritance theme is a chapter from the research
team of William Ballard rooted in research on genetic model Drosophila, in which
paternal mtDNA inheritance is observed much more frequently than in mammals
and under natural conditions rather than by interspecific cross-hybridization or
assisted reproduction (Wolff et al. 2012). Similar to other animal models, UPS and
autophagy contribute to the elimination of paternal mitochondrial genome in Dro-
sophila, when it does occur (Politi et al. 2014). Parent-specific mitochondrial
inheritance variations in the fruit fly family can have a bearing on their metabolism,
fitness, reproduction, and even speciation and evolution. In the third chapter, Wen
C. Aw and coauthors (Aw et al. 2019) review the metabolic aspects of mitochondrial
function and homeostasis and the influence of mtDNA mutation on nuclear–mito-
chondrial cross talk not only in Drosophila but also in a wide range of vertebrate
species, including humans. Also discussed is the balance between mitochondrial
ATP and reactive oxygen species (ROS) production and the sex-biased trade-off
between metabolic efficiency and the cost of gamete production. Authors conclude
that exogenous factors, such as diet, may differentially influence the selective costs
of mtDNA mutations. Further studies of the factors influencing mitochondrial
homeostasis are likely to give valuable insight into evolutionary biology and quan-
titative genetics as well as nutrigenomics and pharmacogenomics.
A gift can sometimes become a curse, quite literally. In the case of mitochondria,
the term mother’s curse has been coined to refer to maternal inheritance of mito-
chondria with suboptimal mitochondrial genomes. Besides human health, the integ-
rity and functionality of mitochondrial genome has a significant bearing on
production traits, fitness, and fertility in economically important livestock species,
affecting both economic and sociological aspects of agriculture worldwide.
Uniquely, large mammals can now be propagated by somatic cell nuclear transfer
(SCNT/cloning) for the purpose of preservation of rare/endangered species and rare
breeds, production of transgenic model animals for biomedical research, and mainly
to accelerate the dissemination of economically desirable livestock genomes with
superior production traits and produce disease-resistant animal strains (Wells and
Prather 2017). In the absence of sperm-specific mitophagy determinants, donor cell
mitochondria and mtDNA are invisible to mitophagic machinery present in the
oocyte cytoplasm, resulting in cloned embryos with heteroplasmy due to the pres-
ence of both donor cell and recipient ooplast mitochondrial genomes. In their
chapter, Kanokwan Srirattana and Justin St. John address mitochondrial genome
inheritance patterns and mitochondrial dysfunction associated with SCNT, focusing
on how they impact the reproductive performance in agriculturally important live-
stock species (Srirattana and St John 2019). Also discussed are the benefits of donor
cell mtDNA depletion and mtDNA supplementation to cloned mammalian embryos,
aimed at enhancing the efficiency of in vitro embryo production. Apart from
technological benefits, the incorporation of mtDNA supplementation into SCNT
protocols offers an intriguing model system for studying mitochondrial inheritance
and mtDNA–nuclear interactions.
viii Preface: A Mother’s Gift

Altogether, the first four chapters solidify the view that sperm mitophagy pathway
is well conserved across vertebrate and invertebrate taxa, sharing common molecular
components while also displaying unique features and variations on the mitophagy
theme (such as the doubly uniparental inheritance in Mytilus) that differentiate it
from autophagy observed in somatic cells. Providing a fitting punctum and an
exclamation mark to this monograph, the closing chapter by Peter Kramer and
Paola Bressan puts mitochondrial health and inheritance in the general context of
human fitness, lifestyle, and reproductive health (Kramer and Bressan 2019). Filled
with thought-provoking ideas and commentaries, this chapter highlights the influ-
ences of mitochondrial health and inheritance on animal fitness and human well-
being. It is intriguing how the mitochondria-produced ROS delicately balance
homeostasis with cellular damage, how prolonged life span may come at the expense
of reproduction (and vice versa), and how mitochondria control immune response,
food intake, and circadian clocks. In authors’ own words, the tiny but mighty
mitochondria are so important that an average human carries 14,000 square meters
of inner mitochondrial membranes, able to generate membrane potential on a par
with the electric field of a lightning bolt. Such a realization sets the stage for
discussing the influence of mitochondrial health and integrity on growth, aging,
reproduction, health and disease, exercise, sleep, diet, and food restriction. The
authors also point out that the ever-so-popular, mitochondria-targeting nutritional
supplements and antioxidants are in fact a double-edged sword. Space is given to the
discussion of mitochondrial involvement in the reproductive process and strategy, as
well as to energy cost of procreation, and to the discussion of assisted reproductive
therapies aimed at enabling conception with less than perfect spermatozoa and eggs
rejuvenated by mitochondrial replacement therapy. Among many lessons learned,
the one close to all of us is that we can live longer, healthier lives simply by offering
only the necessary amount of particular substrates (such as glucose) to our own
mitochondria, by limiting our food intake and our natural urge to store energy in fat
cells. Helpfully, some of the past and current diet fads are discussed in this context.
In a captivating and approachable way, the closing chapter provides a compelling
manifesto for the importance of mitochondrial research for our health, well-being,
and prosperity.
In closing, I would like to thank all contributors and their teams, my fellow
mitochondriacs, for tackling the task of compiling their research with great enthu-
siasm and wit. As an editor, I am very grateful for their insightful contributions and,
with their busy schedules, their willingness to sacrifice their time and
mitochondrion-produced energy to write chapters for this volume. Lastly, I would
like to dedicate this monograph to the memory of my mom, Anna, whom I lost not
long ago and who gave me the best gifts only mothers can give: life, unconditional
love, and healthy mitochondria.

Columbia, MO, USA Peter Sutovsky

Preface: A Mother’s Gift ix

References

Al Rawi S, Louvet-Vallee S, Djeddi A, Sachse M, Culetto E, Hajjar C, Boyd L,

Legouis R, Galy V (2011) Postfertilization autophagy of sperm organelles pre-
vents paternal mitochondrial DNA transmission. Science 334:1144–1147
Al Rawi S, Louvet-Vallee S, Djeddi A, Sachse M, Culetto E, Hajjar C, Boyd L,
Legouis R, Galy V (2012) Allophagy: a macroautophagic process degrading
spermatozoid-inherited organelles. Autophagy 8:421–423
Aw WC, Garvin MR, Ballard JWO (2019) Exogenous factors may differentially
influence the selective costs of mtDNA mutations. Adv Anat Embryol Cell Biol
231:51–74
Kramer P, Bressan P (2019) Mitochondria inspire a lifestyle. Adv Anat Embryol
Cell Biol 231:105–126
Merlet J, Rubio-Pena K, Al Rawi S, Galy V (2019) Autophagosomal sperm organ-
elle clearance and mtDNA inheritance in C. elegans. Adv Anat Embryol Cell Biol
231:1–24
Politi Y, Gal L, Kalifa Y, Ravid L, Elazar Z, Arama E (2014) Paternal mitochondrial
destruction after fertilization is mediated by a common endocytic and autophagic
pathway in Drosophila. Dev Cell 29:305–320
Sato M, Sato K (2011) Degradation of paternal mitochondria by fertilization-
triggered autophagy in C. elegans embryos. Science 334:1141–1144
Song WH, Yi YJ, Sutovsky M, Meyers S, Sutovsky P (2016a) The ART and science
of sperm mitophagy. Autophagy 12:2510–2511
Song WH, Yi YJ, Sutovsky M, Meyers S, Sutovsky P (2016b) Autophagy and
ubiquitin-proteasome system contribute to sperm mitophagy after mammalian
fertilization. Proc Natl Acad Sci U S A 113:E5261–E5270
Srirattana K, St John JC (2019) Transmission of dysfunctional mitochondrial DNA
and its implications for mammalian reproduction. Adv Anat Embryol Cell Biol
231:75–104
Sutovsky P, Song WH (2017) Post-fertilisation sperm mitophagy: the tale of mito-
chondrial Eve and Steve. Reprod Fertil Dev 30:56–63
Sutovsky P, Navara CS, Schatten G (1996) Fate of the sperm mitochondria, and the
incorporation, conversion, and disassembly of the sperm tail structures during
bovine fertilization. Biol Reprod 55:1195–1205
Sutovsky P, Moreno RD, Ramalho-Santos J, Dominko T, Simerly C, Schatten G
(1999) Ubiquitin tag for sperm mitochondria. Nature 402:371–372
Sutovsky P, McCauley TC, Sutovsky M, Day BN (2003) Early degradation of
paternal mitochondria in domestic pig (Sus scrofa) is prevented by selective
proteasomal inhibitors lactacystin and MG132. Biol Reprod 68:1793–1800
Wells KD, Prather RS (2017) Genome-editing technologies to improve research,
reproduction, and production in pigs. Mol Reprod Dev 84:1012–1017
Wolff JN, Nafisinia M, Sutovsky P, Ballard JW (2012) Paternal transmission of
mitochondrial DNA as an integral part of mitochondrial inheritance in
metapopulations of Drosophila simulans. Heredity (Edinb) 110:57–62
Zouros E, Rodakis GC (2019) Doubly uniparental inheritance of mtDNA: an
unappreciated defiance of a general rule. Adv Anat Embryol Cell Biol 231:25–50
Contents

Autophagosomal Sperm Organelle Clearance and mtDNA

Inheritance in C. elegans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Jorge Merlet, Karinna Rubio-Peña, Sara Al Rawi, and Vincent Galy
Doubly Uniparental Inheritance of mtDNA: An Unappreciated
Defiance of a General Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Eleftherios Zouros and George C. Rodakis
Exogenous Factors May Differentially Influence the Selective
Costs of mtDNA Mutations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Wen C. Aw, Michael R. Garvin, and J. William O. Ballard
Transmission of Dysfunctional Mitochondrial DNA and Its
Implications for Mammalian Reproduction . . . . . . . . . . . . . . . . . . . . . . 75
Kanokwan Srirattana and Justin C. St. John
Mitochondria Inspire a Lifestyle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Peter Kramer and Paola Bressan

xi
Autophagosomal Sperm Organelle
Clearance and mtDNA Inheritance
in C. elegans

Jorge Merlet, Karinna Rubio-Peña, Sara Al Rawi, and Vincent Galy

Contents
1 C. elegans mtDNA Inheritance and Fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1 C. elegans to Study Maternal Transmission of mtDNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Sperm Components Entering Embryo at Fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 Sperm Organelles Clearance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1 Autophagy Degradation of MOs and Sperm Mitochondria . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Specific Mechanism of Sperm Mitochondria Degradation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.3 Impact of Oocyte-Derived Mitochondria Dynamics on Sperm Organelles Clearance . . 18
3 Sperm Mitochondria Degradation to Prevent Heteroplasmy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.1 Possible Consequences of Heteroplasmy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2 Probably More than One Mechanism to Prevent mtDNA Transmission . . . . . . . . . . . . . 20
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Abstract The nematode C. elegans represents a powerful experimental system with

key properties and advantages to study the mechanisms underlying mitochondrial
DNA maternal inheritance and paternal components sorting. First, the transmission
is uniparental and maternal as in many animal species; second, at fertilization sperm
cells contain both mitochondria and mtDNA; and third, the worm allows powerful
genetics and cell biology approaches to characterize the mechanisms underlying the
uniparental and maternal transmission of mtDNA. Fertilization of C. elegans oocyte
occurs inside the transparent body when the mature oocyte resumes meiosis I and
passes through the spermatheca. One amoeboid sperm cell fuses with the oocyte and
delivers its whole content. Among the structures entering the embryo, the sperm
mitochondria and a fraction of the nematode-specific membranous organelles are
rapidly degraded, whereas others like centrioles and sperm genomic DNA are
transmitted. In this chapter, we will review the knowledge acquired on sperm
inherited organelles clearance during the recent years using C. elegans.

J. Merlet, K. Rubio-Peña, S. Al Rawi, and V. Galy (*)

Sorbonne Université, CNRS, Institut de Biologie Paris Seine, IBPS, Developmental Biology
Laboratory, UMR7622, Paris, France
e-mail: [email protected]

© Springer Nature Switzerland AG 2019 1

P. Sutovsky (ed.), Cellular and Molecular Basis of Mitochondrial Inheritance,
Advances in Anatomy, Embryology and Cell Biology 231,
https://doi.org/10.1007/102_2018_1, Published online: 23 November 2018
2 J. Merlet et al.

Keywords ALLO-1, Allophagy, Autophagy, C. elegans, LGG-1, LGG-2,

Membranous organelles, Mitophagy, mtDNA inheritance, Sperm mitochondria

1 C. elegans mtDNA Inheritance and Fertilization

1.1 C. elegans to Study Maternal Transmission of mtDNA

C. elegans mitochondrial DNA (mtDNA) is similar in size and gene content to

vertebrates’ mtDNA including human (Anderson et al. 1981). It is a 13.8 kb circular
molecule carrying 36 genes, including 2 ribosomal RNAs, 22 transfer RNAs, and
12 mitochondrial proteins genes (Okimoto et al. 1992). C. elegans germ cells
proliferate and start to differentiate in a syncytium. During oocyte maturation,
cellularization occurs in the proximal part of the gonadal arm of the hermaphrodite
worm (Fig. 1). This cellularization process allows mitochondria incorporation, and
around 25,000 copies of the maternal mtDNA are transmitted to the embryo via the
oocytes’ mitochondria. This copy number remains stable during embryonic and
larval development up to the third larval stage (Tsang and Lemire 2002a). The
first mtDNA amplification, corresponding to a fivefold increase, occurs at the L4
stage, at the time of gonad activation. A second sixfold amplification occurs in the
hermaphrodite adults. This mtDNA amplification is necessary for normal

Fig. 1 C. elegans to study the fate of sperm components after fertilization. Schematic representa-
tion of one of the two gonad arms of a C. elegans hermaphrodite adult. The U-shape syncytial gonad
produces cellularized oocytes arrested in prophase of the first meiotic division. These oocytes
receive maternal mitochondria (green). The most proximal mature oocyte resumes meiosis and is
pushed through the spermatheca for fertilization and toward the uterus where it starts embryonic
development. Upon fertilization and gametes fusion, MOs (blue), sperm mitochondria (red) and
sperm nuclear DNA (violet) enter the oocyte and define the posterior pole of the embryo. Upon
meiosis I and II completion, the eggshell is formed, and the sperm-derived organelles are dispersed
within the cellular volume and randomly segregated between the two blastomers. By the 100-cell
stage (200 min), sperm organelles are degraded
Autophagosomal Sperm Organelle Clearance and mtDNA Inheritance in C. elegans 3

development since inhibition of mtDNA replication by an ethidium bromide treat-

ment induces an L3 arrest (Tsang and Lemire 2002a).
The copy number of mtDNA is sex dependent. Indeed male worms have half of
the hermaphrodite’s mtDNA (Tsang and Lemire 2002a). This difference is even
more spectacular in gametes. The number of mtDNA molecules in C. elegans
spermatozoa was estimated to be around 30–40 (Tsang and Lemire 2003) compared
to the 25,000 copy present in the mature oocyte. Despite their low number, the
mtDNA molecules are visible by fluorescent light microscopy as few bright spots in
the mitochondria of the mature motile spermatozoa cells using SYBR Green, a DNA
intercalant (Fig. 2). At the time of fertilization, sperm mitochondria enter the embryo

Fig. 2 C. elegans spermatozoa contain mitochondria with mtDNA as well as membranous

organelles (MOs). Single spermatozoa observed by fluorescent light microscopy (a and b) or
transmission electron microscopy (c and d). (a) MOs (white arrowheads) were visualized using
the SP56 antibody (green) on a fixed spermatozoon. Sperm mitochondria (red arrowheads) were
labeled using CMXRos (red). Sperm nuclear DNA (asterisk, blue) was labeled with Hoechst. (b)
Sperm mitochondria (red arrowheads) were labeled using tetramethylrhodamine ethyl ester dye
(TMRE, red) on living spermatozoa. Sperm nuclear DNA (asterisk, green) and sperm mitochondrial
DNA (yellow arrowheads) were labeled with SYBR Green (green). (c) Transmission electron
micrograph of a mature spermatozoon after high-pressure freezing and freeze substitution showed
membranous organelles fused with the plasma membrane (white arrowheads) and the sperm
mitochondria (red arrowheads) around the condensed nuclear chromatin (asterisk). (d) At higher
magnification the cristae were visible in the sperm mitochondria. Scale bars are 2.5 μm (a and b),
500 nm (c) and 250 nm (d)
4 J. Merlet et al.

(Al Rawi et al. 2011; Sato and Sato 2011) along with their mtDNA (Zhou et al. 2011,
2016). Despite the 600- to 800-fold dilution of sperm mtDNA compared to the
maternal mtDNA, dilution is not the only mechanism insuring uniparental maternal
transmission of mtDNA (Sect. 2).
The isolation of a mtDNA mutant strain was instrumental to track sperm mtDNA
in the progeny. The most frequently used strain contains the UaDf5 mtDNA allele
which carries a 3 kb deletion affecting 4 proteins-encoding genes and 11 tRNA-
coding genes. Interestingly, this shorter molecule is maintained at a high copy
number in the worms together with wild-type (WT) mtDNA molecules which
complement the not fully functional UaDf5 allele. Therefore, animals carrying
UaDf5 allele are heteroplasmic, with the UaDf5 representing 60% of the total
number of mtDNA molecules which is twice of the total mtDNA molecules in WT
worms (Tsang and Lemire 2002b). Because of its huge size, the UaDf5 deletion can
be easily tracked by PCR. Historically, this allele was used to demonstrate that
mtDNA is maternally transmitted in worms, and then it rapidly became a tool of
choice to study the mechanisms of sperm mtDNA clearance (Al Rawi et al. 2011;
Djeddi et al. 2015; Sato and Sato 2011; Sato et al. 2018; Zhou et al. 2011, 2016).
When heteroplasmic males carrying the UaDf5 allele are crossed with WT hermaph-
rodites, the progeny only contains the WT mtDNA after the 64-cell stage (Zhou et al.
2011). This sperm mtDNA clearance in the embryo to insure strictly maternal
inheritance of mtDNA is correlated with the degradation of labeled sperm-derived
mitochondria (Al Rawi et al. 2011; Sato and Sato 2011; Zhou et al. 2011; Sect. 2).
C. elegans is an androdioecious (male-hermaphrodite) specie with self-fertilizing
hermaphrodites. Neither in self-fertilization nor in cross fertilization, the sperm-
derived mtDNA are kept in the progeny, and no differences in the timing and
mechanism of sperm mitochondria degradation were detected despite their different
origins. In other words, the degradation of the sperm mitochondria and the nematode-
specific Golgi-derived membranous organelles (MOs) occurred also when sperm
mitochondria had the same origin than the oocyte mitochondria. This characteristic
strongly suggests that the mechanisms responsible for sperm mitochondria degradation
recognize sperm property/ies rather than their exogenous origin.
The potential reason for uniparental maternal heredity could be the need to avoid
the transmission of mutated forms of the mtDNA generated during the life of the
motile spermatozoon. The coexistence of different forms of mtDNA molecules,
a.k.a. heteroplasmy, is thought to be deleterious and unstable in the organism
(Sharpley et al. 2012). Even though, the degree of mitochondrial heteroplasmy in
natural worm populations and within single individuals has not been documented,
studies of the heteroplasmic worms carrying the UaDf5 mtDNA allele revealed
defects (Liau et al. 2007; Sato et al. 2018). This heteroplasmic strain is viable,
shows reduced egg-laying and defecation rates, a shorter life-span, as well as a
reduced motility of sperm cells (Liau et al. 2007). Furthermore, the duration of cell
division is extended, and the embryonic lethality is higher in these heteroplasmic
worms with 9.4% of lethality against 0.4% for homoplasmic wild-type worms (Zhou
et al. 2016). However, it is worth noting that this heteroplasmy may represent a very
peculiar case since it is stable over generations (Tsang and Lemire 2002b), while
Autophagosomal Sperm Organelle Clearance and mtDNA Inheritance in C. elegans 5

heteroplasmy in mice has been described as very unstable with the rapid switch to a
homoplasmic status (Sharpley et al. 2012). This stability in the worms carrying the
UaDf5 mtDNA is probably the result of antagonist selection mechanisms. The
smaller mtDNA has a replicative advantage, while the WT mtDNA remains required
for cell/organism viability (Tsang and Lemire 2002b). Interestingly, milder embry-
onic phenotypes (embryonic lethality, cell division duration) were also observed for
embryos showing heteroplasmy due to the transient stabilization of sperm mtDNA
from a different wild-type isolate (Hawaïn crossed with the Bristol isolate) (Zhou
et al. 2016). This experiment demonstrates that transient heteroplasmy induced by a
delayed degradation of sperm mitochondria is deleterious for the embryo. Testing
the impact of heteroplasmy over several generations has not been possible so far
since sperm mitochondria are still degraded later during development, and some of
the mutations or RNAi delaying their degradation, like autophagy inactivation, are
lethal by themselves (Al Rawi et al. 2011; Sato and Sato 2011).

1.2 Sperm Components Entering Embryo at Fertilization

C. elegans spermatozoa are amoeboid cells of around 5 μm. The spermatozoa are
formed at the end of the fourth larval stage in hermaphrodites or in males. Four
spermatids form the budding of one secondary spermatocyte. During the budding,
several cellular components present in the spermatocytes like actin, tubulin, and the
ribosomes are excluded from the forming spermatids and retained in the residual
body (for review, see Nishimura and L’hernault 2017). The nuclear DNA, the MOs,
and the mitochondria are segregated in the four spermatocytes and remain visible in
the mature spermatozoon (Fig. 2). The mature spermatozoa correspond to the
spermatids activated into mobile spermatozoa at the time of the first ovulation or
upon ejaculation of the male with TRY-5 protease in the seminal fluids (Smith and
Stanfield 2011).
In hermaphrodite worms, 99% of the spermatozoa are used to fertilize an oocyte
(Ward and Carrel 1979). When males and hermaphrodites are mated, the spermato-
zoa from the male displace the ones from the hermaphrodite (Ward and Carrel 1979).
At each fertilization, some of the spermatozoa are pushed out of the spermatheca
toward the uterus (Ward and Carrel 1979) and have then to crawl back toward the
spermatheca to get ready for the next fertilization event (Ward and Carrel 1979). The
motility relies on the polymerization activity of the proteins of the family of the
major sperm proteins (MSP) representing up to 40% of the total protein composition
of mature sperm cells (Italiano et al. 1996). MSP assembly into fibers depends on
ATP, and therefore sperm motility requires mitochondria activity for ATP supply.
This production of ATP might generate ROS and represent a threat for sperm
mtDNA integrity. At fertilization, the plasma membrane of the spermatozoon fuses
with the plasma membrane of the oocyte, and its content enters the cytoplasm. This
includes, at least, the MSP proteins, the centrioles, the genomic DNA, the sperm
mitochondria with their mtDNA molecules, and the nematode-specific MOs.
6 J. Merlet et al.

MSP proteins, sperm mitochondria, and MOs disappear soon after fertilization,
while sperm genomic DNA and the centrioles are critical for embryogenesis. In
this chapter, we will focus only in the organelles that are present in the mature
spermatozoa and degraded after their entry.

1.2.1 The Membranous Organelles (MOs)

The membranous organelles (MOs) are nematode-specific sperm organelles with a

bilobed structure composed of a smaller head that is separated by an electron-dense
collar from a larger body (Fig. 2). MOs are Golgi-derived vesicles that fuse with the
plasma membrane during the maturation of spermatocytes into mature spermatozoa.
During spermatogenesis MOs are associated with the fibrous body (FB) and form the
FB-MO complex involved in partitioning the cytoplasm during spermatogenesis.
MOs also play a critical role during the asymmetric cell division occurring in sperm
meiosis II (Ward et al. 1981; Wolf et al. 1978). One major role of the FB-MO
complex is to ensure the proper segregation of proteins into the spermatids rather
than to the residual body, and during spermatozoa maturation, new components are
added to the cell surface by MO fusion to the plasma membrane (Chatterjee et al.
2005; Roberts et al. 1986; Shakes and Ward 1989; Ward et al. 1981; Xu and
Sternberg 2003). Therefore, MO morphology changes dramatically during sper-
matogenesis, and blocking MO fusion results in non-motile spermatozoa unable to
adhere to the uterine walls (Machaca and L’hernault 1997; Washington and Ward
2006).
Even when most MOs are fusing upon sperm maturation, a fraction remains
unfused as well as the body of the MOs (Nelson and Ward 1980) and still contains at
least the SP56 antigens. So, some MOs enter the ooplasm at fertilization and are
specifically targeted for degradation. One could speculate that MO membranes and
associated proteins could be harmful for the embryo. For example, MO could fuse
with membranous compartments in the oocyte and alter their functions. Indeed,
FER-1, which contains six C2 domains (protein kinase C conserved region 2)
domains which usually act in calcium-dependent lipid processing events, such as
vesicle fusion (Lemmon 2008), is one of the identified proteins localized in MOs
(Washington and Ward 2006). FER-1 regulates the calcium-dependent membrane
fusion between MOs and the plasma membrane. Therefore, after fertilization and the
entry of unfused MO into the cytoplasm of the oocyte, the degradation of the MOs
would be a mechanism to get rid of FER-1 and avoid their inappropriate fusion with
embryo vesicles.

1.2.2 Sperm Mitochondria

Mitochondria in C. elegans spermatozoa are distributed around the condensed

chromatin and excluded, like the MOs, from the pseudopod region. TEM analysis
revealed that they tend to have circular sections, while the mitochondria of maternal
Autophagosomal Sperm Organelle Clearance and mtDNA Inheritance in C. elegans 7

origin in the embryo show sections of elongated shapes. The average diameter of the
wild-type sperm mitochondria is around 460 nm which makes them distinguishable
from the maternal ones which are more tubular with an average diameter of around
240 nm (Zhou et al. 2016).
The observation of the mitochondria morphology by light microscopy shows
clear differences between oocyte and sperm-derived mitochondria within the 1-cell
stage embryo completing meiosis division (Al Rawi et al. 2011; Sato and Sato 2011;
Fig. 3). The few sperm mitochondria are localized around the condensed sperm
chromatin at the posterior side of the embryo (Fig. 3), and they appear more
fragmented than the elongated and interconnected maternal mitochondria (Fig. 4).
The sperm mitochondria show the presence of clear cristae by TEM and are labeled
by fluorescent dyes accumulating into mitochondria with a membrane potential
arguing for intact, functional, and polarized sperm mitochondria prior fertilization
(Al Rawi et al. 2011; Sato and Sato 2011).

2 Sperm Organelles Clearance

2.1 Autophagy Degradation of MOs and Sperm Mitochondria

2.1.1 Kinetic of Sperm Organelles Clearance

One remarkable property of the process in C. elegans compared to other species is its
speed. Sperm mitochondria and MOs are rapidly degraded in the embryo before the
64-cell stage in less than 3 h of development (Al Rawi et al. 2011; Sato and Sato
2011). By comparison sperm mitochondria are degraded by 84 h in the mouse
embryo (Luo et al. 2013; Rojansky et al. 2016). In Drosophila, the sperm mtDNA
is eliminated prior to fertilization, and the mitochondria are degraded in the embryo
within few hours (Politi et al. 2014).
The LGG-1 and LGG-2 autophagy proteins (C. elegans homologs of LC3/Atg8
proteins) are recruited around MOs and sperm mitochondria at the end of the first
female meiosis division (15–20 min post fertilization) (Al Rawi et al. 2011; Sato and
Sato 2011; Fig. 5). LGG-1 is required for autophagosome formation and LGG-2
recruitment, while LGG-2 is dispensable. Interestingly, these respective contribu-
tions in the formation of autophagosomes are correlated with the impact of their
depletion on embryo development. LGG-1 is essential, while worms lacking LGG-2
are viable (Djeddi et al. 2015; Manil-Segalen et al. 2014). This suggests that their
respective function in sperm-inherited organelle degradation is characteristic of their
function in other autophagic processes during development.
In the 1-cell stage embryo, the localization of the autophagy markers is restricted
to the area around sperm-inherited organelles. LGG-1 is localized in few cytoplas-
mic foci in the oocytes (Al Rawi et al. 2011; Sato and Sato 2011). After fertilization
LGG-1 and LGG-2 are essentially localized around sperm organelles at the posterior
pole of 1-cell embryos ongoing first and second meiotic divisions (Al Rawi et al.
8 J. Merlet et al.

Fig. 3 MOs and sperm mitochondria are found in the cytoplasm, at the posterior pole of the 1-cell
stage C. elegans embryo shortly after fertilization. (a) MOs and the sperm mitochondria were
observed in spermatozoa outside the embryo and inside a 1-cell stage embryo by confocal
microscopy. MOs were visualized using the SP56 antibody (green) on a fixed sample, sperm
mitochondria were labeled using CMXRos (red), and nuclear DNA was labeled with Hoechst
(blue). At the end of the second meiotic division (around 30 min after fertilization), MOs and sperm
mitochondria are still grouped around the sperm nuclear DNA, at the posterior pole of the 1-cell
embryo, ready to be targeted by the autophagy machinery. The dotted line indicates the border of
the embryo; sperm- (♂) and oocyte-derived (♀) nuclear DNA are indicated. (b and c) Transmission
electron micrograph of spermatozoid organelles around sperm nuclear chromatin (asterisk) in a
1-cell embryo at metapase of the first meiosis division, revealing the presence of (c) sperm-derived
mitochondria (♂mito) with granules in their matrix and (b) membranous organelles (MO). Scale
bars are 2.5 μm in (a) and 500 nm in (b and c)

2011; Sato and Sato 2011). Therefore, in C. elegans, autophagy degradation of

sperm organelles represents the main ongoing autophagy at this time. This is
different to the crowded autophagy pattern observed in early developing mouse
Autophagosomal Sperm Organelle Clearance and mtDNA Inheritance in C. elegans 9

Fig. 4 Following fertilization, C. elegans sperm mitochondria are outnumbered by the oocyte-
derived mitochondria and show a distinct morphology. Oocyte-derived mitochondria labeled with
GFP::ANT-1 protein (green) and sperm-derived mitochondria labeled with CMXRos mitotracker
(red) were visualized in a fixed 1-cell stage embryo by fluorescent microscopy. DNA was labeled
with Hoechst (blue). Scale bar represents 5 μm

embryos where autophagy has been proposed to degrade maternal proteins

(Tsukamoto et al. 2008). Along the same line, inactivation of autophagy (conditional
KO of Atg5) in mouse embryo prevents the development beyond the 8-cell stage
(Tsukamoto et al. 2008). In C. elegans, lgg-1 and lgg-2 RNAi-treated hermaphro-
dites produce a progeny showing 80% of late embryonic lethality and 20% of L1
arrest (Al Rawi et al. 2011; Sato and Sato 2011, VG unpublished results). lgg-1 is a
maternal effect gene: the embryonic and larval development are possible due to the
maternal contribution from the heterozygote mother. Indeed, lgg-1 homozygote
mutant embryos develop into sterile adults. Therefore, bulk autophagy is, like in
vertebrates, an essential process for worm development (Palmisano and Melendez
2018). On the other hand, it was possible to demonstrate that autophagy and the
lysosomal pathway are specifically involved in sperm organelle clearance since
sperm organelle autophagy is the most prominent ongoing autophagy process in
the first 3 h after fertilization. Indeed the inactivation of these pathways using
mutants or RNAi depletions allowed the stabilization of sperm mitochondria and
their genome as well as of the MOs (Al Rawi et al. 2011; Sato and Sato 2011; Zhou
et al. 2011; Fig. 6), while it has no stabilization effect on the degradation of a family
of very abundant sperm proteins that enter the embryos at fertilization, the MSP
proteins (Al Rawi et al. 2011). The MSP proteins are degraded between the 2- and
4-cell stage by an autophagy-independent process (Al Rawi et al. 2011).
10 J. Merlet et al.

Fig. 5 Autophagy markers are recruited around sperm mitochondria at the posterior pole of
C. elegans 1-cell stage embryo. LGG-1 (a) and LGG-2 (b) labeled using specific antibodies
(green) in a fixed 1-cell stage embryo were visualized by fluorescent microscopy. Sperm-derived
mitochondria were labeled with CMXRos mitotracker (red), and DNA was labeled with Hoechst
(blue). Scale bar represents 5 μm

2.1.2 Autophagosome Formation, Maturation, and Dynamics

In WT embryos, the sperm-inherited organelles are initially clustered around the

sperm nuclear chromatin at the posterior pole of the embryo (Fig. 3). The autophagy
induction depends on the entry of sperm components in the embryo. This was
demonstrated using a spe-9 fertilizing mutant. The spe-9(hc52ts) is a thermosensitive
mutant that prevents the fusion of the spermatozoa with the oocyte. At restrictive
temperature, it accumulates unfertilized oocytes in the uterus, and no GFP::LGG-1
accumulation was observed in these oocytes (Sato and Sato 2011). On the other
hand, oocyte fertilization by two spermatozoa induced upon egg-4/5 (RNAi) deple-
tion triggers the localization of endogenous LGG-1 protein around the sperm
organelles of both spermatozoa (Al Rawi et al. 2011). egg-4/5(RNAi) polyspermic
embryos also allowed to demonstrate that the autophagy response is independent of
Autophagosomal Sperm Organelle Clearance and mtDNA Inheritance in C. elegans 11

Fig. 6 Autophagy is required to degrade C. elegans spermatozoid-inherited organelles after

fertilization and prevents mitochondrial heteroplasmy. (a) In 120-cell embryos, the MOs (green)
are absent in control (left) but remain in lgg-1(tm3489) mutant embryo. LGG-1 (red) is present in
WT but not in lgg-1(tm3489). Pictures are Z-projections of confocal stacks. Paternal mitochondria
(green) from labeled males are absent in control (left) but remain in lgg-1 + 2(RNAi) embryo. Scale
bar are 10 μm. (b) Interfering with autophagy maintains paternal mitochondrial heteroplasmy.
Heteroplasmic males carrying both deleted (UaDf5) and WT mtDNA were crossed with either
control or lgg-1 + 2(RNAi) hermaphrodites, and their embryos progeny were tested by PCR for the
presence of wild-type (WT) and mutated (UaDf5) mtDNA. Male UaDf5 mtDNA is detected in
autophagy-deficient embryos, but not in control, indicating an heteroplasmic state. Adapted with
permission from Al Rawi et al. (2011)

the posterior position of the substrates (Al Rawi et al. 2011). The same double
induction of autophagy was also observed in polyspermic embryos obtained in the
spe-11(hc77ts) embryos (Sato and Sato 2011), a protein not required for fertilization
but for embryo development. The induction of this early autophagy is also indepen-
dent of cell cycle progression since, in arrested embryos at metaphase of meiosis I
upon emb-27(RNAi) treatment, GFP::LGG-1 was still recruited around the sperm
organelles (Sato and Sato 2011).
Electron tomographs revealed the modification of sperm mitochondria shortly
after their entry in the embryos. The cristae gradually lose their structure, and
electron-dense granules appear in the matrix (Zhou et al. 2016; Fig. 3). These
types of granules were observed previously (Fawcett 1981) and their origin and
potential function discussed (Jacob et al. 1994). They are probably composed of
phospholipids and able to bind calcium, but their biological functions are still
unknown. These aggregates are growing in size, while the cristae are cleared prior
12 J. Merlet et al.

to autophagosome enclosure. This made the authors suggest and test the hypothesis
that the paternal elimination begins as a self-destruction process. A dramatic change
in morphology and ions exchanges in sperm mitochondria at the time of fertilization
has been described in acidians (Lambert and Epel 1979). In these marine species
when sperm attach to the chorion, it triggers the swelling of the mitochondrion and
its physical exclusion from the spermatozoon along the flagellum. This swelling is
coupled with the loss of protons from the mitochondria and sperm cell. This reaction
can be triggered by increasing the pH or lowering the Na+ of the sea water. One
could speculate that the same type of reaction is triggered on C. elegans sperm
mitochondria upon their entry.
The autophagy of the sperm-inherited organelles is spontaneous and physiolog-
ical, making it an attractive model to study macro-autophagy. It has been used to
decipher the specific functions of LGG-1 and LGG-2 in the allophagosome forma-
tion and maturation. LGG-1 and LGG-2 are both localized around sperm organelles
but do not completely overlap (Djeddi et al. 2015). lgg-2 RNAi depletion reduced
the viability of the dauer (a stage of developmental arrest) worms and the lifetime of
the adults (Alberti et al. 2010). lgg-2 RNAi treatment was lethal in mutant carrying a
loss-of-function mutation in daf-2, the C. elegans insulin-like tyrosine kinase recep-
tor, which triggers abnormal constitutive dauer entry (Melendez et al. 2003). LGG-2
function in autophagy was further supported since lgg-2 RNAi depleted embryos
abnormally accumulate P-granules components in the somatic cells (Zhang et al.
2009). The characterization of the lgg-2(tm5755) / embryos suggested that
LGG-2 is required for allophagosomes acidification, a function that would be
mediated by an interaction with VPS-39 (Manil-Segalen et al. 2014), one component
of the HOPS complex, a multimeric tethering protein complex involved in vesicle
fusion of late endosomes. In the absence of LGG-2 protein, LGG-1 was still
recruited around the substrates that appeared clustered in the early embryos (Djeddi
et al. 2015; Manil-Segalen et al. 2014). This clustering phenotype of the
allophagosomes was proposed to be the consequence of a defect in allophagosomes
acidification due to the lack of fusion with the lysosomes as a result of the loss of
interaction with VPS-39 (Manil-Segalen et al. 2014). The potential link with the
endosomal pathway was intriguing and could be conserved among species as
suggested by the localization of the VPS-27 in potential hybrid compartments
between endosomes and autophagosomes (amphisomes) around sperm-inherited
organelles in the 1-cell stage C. elegans embryos (Manil-Segalen et al. 2014) as
well as the colocalization of Rab7 with sperm mitochondria in Drosophila embryos
(Politi et al. 2014).
Live embryo imaging of labeled sperm mitochondria and GFP::LGG-1 allowed
to describe the dynamics and distribution of the allophagosomes during the first hour
of C. elegans development. The allophagomes are formed around the clustered
sperm organelles near the condensed sperm nuclear DNA at the posterior pole of
the embryo. The substrates and the autophagosomes tend to migrate toward the
anterior pole of the embryos at the time of pronuclei formation. After pronuclei
meeting and during pronuclei centration, the autophagosomes tend to be gathered by
the active centrosomes. During the first mitosis, they are randomly distributed
Autophagosomal Sperm Organelle Clearance and mtDNA Inheritance in C. elegans 13

between the two daughter blastomers (Djeddi et al. 2015; Hajjar et al. 2014). In the
absence of LGG-2, the dynamics of the allophagosomes was dramatically impaired.
LGG-1-labeled autophagosomes remained clustered close to the 1-cell stage plasma
membrane, and their dispersion between P1 and AB blastomers was abnormal
(Djeddi et al. 2015). This defect in the intracellular dynamics was correlated with
a delay in the degradation of the sperm-derived MOs and mitochondria (Djeddi et al.
2015). This extended clustering phenotype is also coupled to a block in the retro-
grade movement of the autophagosomes toward the pericentrosomal area where the
acidic compartment tends to accumulate. This suggests that the delay in sperm-
derived organelles clearance could be due to a delay in the fusion of autophagosomes
with acidic compartments (Djeddi et al. 2015).

2.1.3 Common Recognition Signal of Sperm Organelles

ALLO-1/IKKE-1

The recent identification of IKKE-1, the worm homologue of the TBK1/IKKε

protein kinase family, and ALLO-1, an autophagy receptor, provides important
information on the mechanism of induction of sperm organelle-specific autophagy
(Fig. 7). These factors are expressed in the germ line and required for normal
elimination of MOs and sperm mitochondria in early embryogenesis, before 8- or
16-cell stage (Sato et al. 2018). The ALLO-1 protein is maternally contributed and
recruited around sperm-inherited organelles within 10 min post fertilization. This is
around 5 min before the recruitment of LGG-1 protein, an autophagy marker
(Al Rawi et al. 2011; Sato and Sato 2011). The recruitment still occurs even upon
the depletion of ATG-11, an early autophagy factor (Sato et al. 2018). ALLO-1
recruitment on the autophagy substrates is therefore a very early step in the process,
independent of the autophagy machinery. ALLO-1 contains an N-terminal
LC3-interacting region (LIR) responsible for its binding to LGG-1 but dispensable
for its localization on the substrates (Sato et al. 2018).
Recognition of ALLO-1 by LGG-1 required ALLO-1 phosphorylation by
IKKE-1, but other kinase(s) might be involved. Indeed, a GFP::IKKE-1 kinase
dead mutant still localizes to the sperm organelles in the embryo but delays their
degradation (Sato et al. 2018). Even if IKKE-1 is not the only kinase responsible for
ALLO-1 phosphorylation, it participates in its regulation (Sato et al. 2018).

Ubiquitination of the Sperm Organelles

Prior to fertilization, MOs are ubiquitinated, as shown by immunofluorescence

where anti-ubiquitin antibodies decorate MOs in the sperm (Al Rawi et al. 2011;
Sato and Sato 2011). After fertilization, a new wave of branched poly-ubiquitin
deposition recognized by K48- and K63-specific antibody arises (Al Rawi et al.
2011; Hajjar et al. 2014).
14 J. Merlet et al.

Fig. 7 The mechanisms involved in sperm organelles autophagy degradation. After the fusion of
the spermatozoon and the oocyte, the sperm mitochondria and the MOs are targeted by the
autophagy machinery with the recruitment of the membrane-associated LGG-1 protein. Several
structural modifications and mechanisms were recently described, but more work is required to
evaluate their respective contribution and relationship. In all pathways, sperm mitochondria lose
their membrane potential (not represented) and accumulate in their matrix electron-dense granules
(black dots) of unknown origin and function. (Left) The sperm provided endonuclease G (CSP-6,
blue) is relocalized in the matrix after fertilization to degrade mtDNA (red circles). This contributes
to sperm mtDNA clearance and sperm mitochondria cristae destabilization before autophagosome
Autophagosomal Sperm Organelle Clearance and mtDNA Inheritance in C. elegans 15

In vertebrates, the ubiquitin mark on sperm mitochondria has been described

(Sutovsky et al. 1999). Therefore, the role of the ubiquitin modifications for sperm
mitochondria recognition and degradation in C. elegans has been considered but
then rejected based on the absence of obvious ubiquitin marks on fixed samples
(Al Rawi et al. 2011; Hajjar et al. 2014; Sato and Sato 2011), the colocalization of
GFP::ubiquitin with membranous organelles, and the absence of clear association
with the sperm mitochondria (Al Rawi et al. 2011). However, this has been recently
revisited (Sato et al. 2018; Fig. 7). Using GFP-specific antibodies, traces of GFP::
ubiquitin signal were found to be associated with sperm mitochondria after their
entry suggesting that sperm mitochondria may contain a low level of ubiquitinated
proteins (Sato et al. 2018). In addition, upon RNAi depletion of UBA-1, the unique
E1-ubiquitin-conjugating enzyme in worms, the recruitment of autophagy adaptor
protein ALLO-1 was prevented (Sato et al. 2018). Ubiquitination was therefore
proposed as a signal for ALLO-1-dependent autophagy on MOs as well as sperm
mitochondria (Sato et al. 2018). While ubiquitin marks already decorate MOs in
activated sperm cells, it is not known whether sperm mitochondria are already
marked before fertilization (Sato et al. 2018) and if ubiquitin serves as the first
mark for sperm mitochondria targeting. Interestingly proteasome subunits were
visualized in spermatozoa and around sperm organelles after their entry (Al Rawi
et al. 2011; Hajjar et al. 2014). With two different fluorescent tags allowing to
visualize the sperm and the oocyte contributed proteasomal 19S subunits RPT-1,
different localizations emerged: the maternally provided RPT-1 localized on MOs
while the sperm-contributed fraction around sperm organelles (Hajjar et al. 2014).
This dual localization suggests dual function of the proteasome in sperm organelle
clearance, but more work is required to identify the potential substrate(s) (Hajjar
et al. 2014). These observations were done using RPT-1 subunits expressed under a
germ line-specific promotor. It would be interesting to confirm these results using
tagged versions of several proteasome subunits expressed under their endogenous
promotors using CRISPR-Cas9 genome editing. Interestingly, in Hela cells, LC3 is
recruited to the inner mitochondrial membrane via its interaction with the
LC3-interacting region (LIR) of the mitochondrial protein PHB2, and this occurs
at sites of proteasome-dependent ruptures of the outer mitochondrial membrane
(Wei et al. 2017), a process that might depend on ubiquitination. The role of
ubiquitination of sperm mitochondria proteins in C. elegans has to be confirmed.

⁄


Fig. 7 (continued) formation. (Center) Maternal prohibitin 2 (PHB-2), an internal mitochondrial

membrane protein, is exposed to the cytosol due to the outer membrane rupture. PHB-2 might
participate in LGG-1 recruitment and autophagosome formation. (Right) Sperm mitochondria and
MOs are ubiquitinated in the ooplasm and recruit the ALLO-1 autophagy adaptor (hexagon) and its
kinase IKKE-1 (circle). IKKE-1 phosphorylates ALLO-1, and this contributes to autophagosome
assembly and sperm organelle degradation. ALLO-1 recruits LGG-1 and associated membranes via
its LIR domain. CSP-6 and PHB-2 are specific of the sperm-derived mitochondria, while ALLO-1/
IKKE-1 are involved in the targeting of both the MOs and the sperm-derived mitochondria
16 J. Merlet et al.

2.2 Specific Mechanism of Sperm Mitochondria Degradation

2.2.1 Loss of Mitochondrial Membrane Potential

A common signal in mitophagy is the loss of membrane potential. This can be

evaluated using tetramethylrhodamine ethyl ester (TMRE), a fluorescent dye only
imported in mitochondria with a membrane potential. Therefore, the hypothesis that
sperm mitochondria lose their membrane potential was evaluated. A loss of mem-
brane potential based on the loss of TMRE signal from sperm mitochondria after
their entry into the embryo was observed (Sato et al. 2018; Zhou et al. 2016).
However, no direct observation of this loss of TMRE staining has been observed
in utero. Interestingly, the loss of membrane potential was also observed in mouse
embryo between 18 and 48 h of development suggesting that the loss of membrane
potential might be a common event in sperm mitochondria degradation. Even if the
sequence appeared conserved, the timing is different since the loss of membrane
potential seems to occur in less than 10 min in C. elegans (Sato et al. 2018), while it
takes more than 18 h in mice (Rojansky et al. 2016). It must be noticed that there are
yet no direct experimental evidences that the loss of membrane potential has a
functional role in sperm mitochondria degradation (Rojansky et al. 2016).
When males were labeled using TMRE and SYTO11 and crossed with unlabeled
hermaphrodites, no TMRE signal was colocalized with the SYTO11 signal in the
1-cell stage embryo (Zhou et al. 2016) suggesting the absence of TMRE signal in
sperm mitochondria after their entry in the unlabeled oocytes (Sato et al. 2018; Zhou
et al. 2016; our unpublished data). The absence of TMRE in sperm mitochondria in
TMRE-labeled embryos was also observed when SYTO11-labeled males were
crossed with TMRE-labeled hermaphrodites. In this last experiment, only the
oocyte-derived mitochondria were labeled with TMRE (Zhou et al. 2016). This
result would argue for a loss of membrane potential of sperm mitochondria in the
embryo rather than the dilution effect of the TMRE in an unlabeled embryonic
environment. More experiments are needed to confirm this interpretation and to
clarify the potential causal relationship between an early loss of membrane potential
and their targeting by autophagy. It is also not clear yet what would be the
mechanism responsible for the loss of the membrane potential. The loss of TMRE
from the sperm-derived mitochondria was observed even in allo-1 and ikke-1 mutant
embryos. This argues for the loss of potential being a very early step not depending
on ALLO-1 recruitment nor IKKE-1 kinase activity (Sato et al. 2018).

2.2.2 Endonuclease mtDNA Degradation

Interestingly, in C. elegans, like many animal species, the 35–50 copies of sperm
mtDNA (Al Rawi et al. 2011) are largely outnumbered in the embryo containing
around 2.4  104 copies of maternal mtDNA (Tsang and Lemire 2002a). Despite the
strong dilution of sperm mtDNA molecules, an active mechanism is required to
degrade them in the embryo.
Autophagosomal Sperm Organelle Clearance and mtDNA Inheritance in C. elegans 17

The mtDNA clearance in other animal species has been described before and after
fertilization (Nishimura et al. 2006). In the fruit fly, it requires the activity of endoG,
an endonuclease protein contributing to the degradation of mtDNA during sperma-
tozoid individualization (Politi et al. 2014). This is coupled to a mechanical exclu-
sion of mtDNA nucleoids during the individualization of the flagella. In endoG
mutant flies, the mtDNA is stabilized.
In C. elegans, an RNAi screen against predicted mitochondrial proteins encoded
by the nuclear genome coupled to a PCR-based assay to detect mtDNA stabilization
revealed CSP-6, a mitochondrial endonuclease G, as a protein required for normal
timing of sperm mtDNA clearance in the embryos (Fig. 7). These results were
confirmed in a mutant expressing a CSP-6 with a deletion of its catalytic domain
(Zhou et al. 2016). In embryos lacking a functional CSP-6 protein, a trackable form
of sperm mitochondrial genome was found beyond the 64-cell stage up to the
fourfold stage of embryo development (around 11 h of development). The observa-
tion of embryos from CSP-6-depleted hermaphrodite crossed with WT males with
mitotracker-labeled mitochondria revealed that sperm mitochondria remain visible
up to the coma stage (around 7 h of development). The apparent shift in stability
between the sperm mtDNA and the labeled sperm mitochondria may be caused by
several experimental differences with the PCR assay being a more sensitive technic
to reveal residual sperm mitochondria material. The delay in sperm mtDNA and
mitochondria degradation upon CPS-6 loss was associated with a slight delay in the
autophagosomes enclosure around sperm mitochondria from the 4- to the 16-cell
stages (Zhou et al. 2016).

2.2.3 Prohibitin 2 in Sperm Mitochondria Degradation

A sperm factor that could serve as a mark for sperm mitochondria targeting has been
proposed (Fig. 7). The worm homologue of vertebrate prohibitin 2 (PHB2) mito-
chondrial protein is required for normal sperm mitochondria degradation (Wei et al.
2017). RNAi depletion of PHB 2 in adult males prior to their crosses delays sperm
mitochondria degradation in the embryos. Sperm mtDNA was also stabilized in the
F1 progeny from the cross involving phb-2 RNAi-treated males (Wei et al. 2017).
Interestingly, a proteasome-dependent rupture of the outer mitochondrial membrane
is required for the mitophagy in HeLa cells and the interaction of LC3 with PHB2.
Surprisingly, while PARKIN and MUL1 were shown to be required for sperm-
derived mitochondria mitophagy in mouse embryos (Rojansky et al. 2016), inacti-
vation of the Pink/Parkin pathway in C. elegans did not prevent sperm mitochondria
degradation. Indeed, paternal mitochondria labeled with HSP-6::GFP and MOs were
still degraded in the sqst-1 (p62 homologue), pink-1, and pdr-1 (Parkin homologue)
deletion mutants (Sato et al. 2018). This result indicates that the mechanism of
degradation of sperm-inherited mitochondria is different from the classical
mitophagy pathway targeting defective mitochondria. This is surprising since
PHB2 is required for Parkin-mediated mitophagy in murine embryonic fibroblasts.
More work is required to understand how phb-2(RNAi) depletion in sperm cells
stabilizes the sperm mitochondria in the C. elegans embryo.
18 J. Merlet et al.

2.3 Impact of Oocyte-Derived Mitochondria Dynamics

on Sperm Organelles Clearance

It has been suggested that mitochondria dynamics, this means a proper balance of
fusion and fission of both maternal and paternal mitochondria, has an important
effect in sperm organelles clearance after fertilization. Mitochondria continually
change shape through the combined actions of fusion and fission (van der Bliek
et al. 2013). In the case of mitochondrial fusion, tubular and elongated organelles are
generated, while fission process generates fragmented ones. C. elegans fzo-1 and
drp-1 genes encode GTPases of the dynamin family that are orthologues of the
MFN1/FZO1 protein (Eura et al. 2003; Rolland et al. 2009), and the DRP1 protein
(Labrousse et al. 1999), which are required for fusion and fission of mitochondria,
respectively (Zamponi et al. 2018).
Loss-of-function mutations in these genes, drp-1(tm1108) which causes severe
mitochondrial fission defects and fzo-1(tm1133) which affects normal fusion dynam-
ics of mitochondria, allowed the study of mitochondrial network and morphology in
C. elegans embryos and sperm (Wang et al. 2016). Using TMRE to specifically label
mitochondria and electron microscopy analysis, Wang et al. showed that mitochon-
dria in WT embryos display different elongated shapes and sizes and are broadly
distributed in the cytoplasm. As anticipated, in mutants affecting mitochondrial
dynamics, maternal mitochondria show radically different morphology and distri-
bution. In drp-1(tm1108) embryos, asymmetric concentrated clusters of long and
highly connected mitochondria are observed, while in fzo-1(tm1133) embryos mito-
chondria are mostly fragmented. Interestingly, in the fzo-1(tm1133); dpr-1(tm1108)
double mutant, fusion defects caused by fzo-1(tm1133) is completely suppressed by
the mitochondrial fission defect caused by drp-1(tm1108) leading to the formation of
long and highly connected mitochondrial. Moreover, each mutation delayed the
clearance of sperm mitochondria labeled with Mitotracker Red (MTR) in cross-
fertilized embryos from WT males with drp-1(tm1108) as well as WT males with
fzo-1(tm1133) hermaphrodites. This demonstrates that mitochondrial dynamics
equilibrium in the embryo is important for paternal mitochondria elimination in
C. elegans.
The impact of the mutations affecting mitochondrial dynamics was also tested on
sperm mitochondria. While EM analysis of WT worms revealed mostly spherical
mitochondria with a diameter of 0.5 μm, they appear larger in fzo-1(tm1133) mutant
and larger but elongated in drp-1(tm1108) mutant spermatozoa. The authors observed
that the paternal fission defect in drp-1(tm1108) males led to the stabilization of the
sperm mitochondria up to the 500-cell stage, while sperm mitochondria from the fzo-1
(tm1133) males were destabilized with a total removal of these organelles at the 32-cell
stage. In cross-fertilized embryos of fzo-1(tm1133) and drp-1(tm1108) worms, these
opposite effects neutralize each other, and mitochondria are eliminated at a normal rate
and are cleared in the 64-cell stage just as in the control.
The stabilization of sperm mitochondria in embryos coming from WT
MTR-stained males and fzo-1(tm1133) hermaphrodites was consistent with the
Autophagosomal Sperm Organelle Clearance and mtDNA Inheritance in C. elegans 19

lack of autophagosome formation around sperm mitochondria in these zygotes. The

quantification of the colocalization of immunolocalized LGG-1 and MTR-labeled
paternal mitochondria revealed that fusion defects from the maternal side impair
autophagosome formation around sperm mitochondria. Interestingly, once again this
impairment is overcame in the fzo-1; drp-1 double mutant hermaphrodite.
Additional TMRE and TEM experiments showed that fzo-1(tm1133) maternal
mitochondria were targeted by the autophagy machinery, which led to think that a
competition between maternal and paternal mitochondria for the autophagy machin-
ery could be responsible for the delay of sperm organelles clearance in these zygotes.

3 Sperm Mitochondria Degradation to Prevent

Heteroplasmy
3.1 Possible Consequences of Heteroplasmy

The strain containing the steady-state heteroplasmy with the UaDf5- truncated allele
of the mtDNA was instrumental, once again, to evaluate the consequences of
heteroplasmy brought by the spermatozoon upon fertilization (Tsang and Lemire
2002b). As mentioned in Sect. 1.1, this viable heteroplasmic strain shows reduced
egg-laying and defecation rates, a shorter life-span, as well as a reduced motility of
sperm cells (Liau et al. 2007). It also displays a higher embryonic lethality than
homoplasmic wild-type worms (Zhou et al. 2016). In previous experiments, the
stabilization of sperm mitochondria was achieved by bulk autophagy inactivation
(Al Rawi et al. 2011; Sato and Sato 2011; Zhou et al. 2011) and lysosomal defects
(Zhou et al. 2011), two cellular functions essential for worm development that are
involved in many aspects of embryonic development (Palmisano and Melendez
2018). Furthermore, none of them is specifically interfering with the sole degrada-
tion of sperm mitochondria since they also delay the clearance of the MOs (Al Rawi
et al. 2011; Sato and Sato 2011). It was therefore difficult to evaluate the impact on
the progeny of the stabilized sperm mitochondria.
Interestingly, the delay in sperm mtDNA clearance has been, for the first time,
associated with a detectable phenotype in the heteroplasmic embryos. A first cross
experiment was done using heteroplasmic males containing sperm mitochondria
with both the UaDf5 and WT haplotypes (Zhou et al. 2016). Crossing these
heteroplasmic males with CPS-6-deficient hermaphrodites produced a progeny
with a higher level of embryonic lethality compared to the crosses using
homoplasmic WT males (5.9 vs 0.7% embryonic lethality). The cell division timing
of embryonic blastomers was doubled in crosses with the heteroplasmic males
compared to the crosses using homoplasmic WT males. Significant but milder
defects were also observed for the progeny of crosses between two C. elegans
haplotypes from WT isolates when the males carrying the N2 haplotype were
defective for CPS-6 activity (Zhou et al. 2016). This is the first indication that
20 J. Merlet et al.

embryonic heteroplasmy induced by the stabilization of sperm-inherited mitochon-

dria has deleterious effects with an increased lethality and cell cycle duration. It
would be interesting to conduct the same phenotypical analysis in embryos with
stabilized sperm mitochondria by other experimental means, for example, upon
inactivation of the autophagy adaptor protein ALLO-1 (Sato et al. 2018). allo-1
mutants are viable (Sato et al. 2018); therefore, it would be interesting to measure
cell cycle duration as well as embryonic lethality.

3.2 Probably More than One Mechanism to Prevent mtDNA

Transmission

Sperm mtDNA and mitochondria may be stabilized in the F1 progeny by several

experimental means (see above). None of the methods used so far allowed obtaining
a stable and inherited heteroplasmy. This is surprising since the trackable sperm
mtDNA used was the deleted UaDf5, a molecule initially obtained after a mutagen-
esis, proving that it was able to colonize the worm. Nevertheless, since the total
mtDNA content is stable until the third larval stage and the increase in mtDNA
quantity is linked to the increase in the number of germ cells, it is likely that there is
no mitochondrial DNA replication and this is therefore not a favorable cellular
context to allow the amplification of the sperm-derived mtDNA. The sperm-derived
mtDNA must reach the germ line lineage during embryonic development to get a
chance to be amplified. In such situation, the smaller UaDf5 mtDNA would have
then a replicative advantage compared to the WT maternal mtDNA and might invade
the germ line and be transmitted to the progeny. When the sperm mitochondria
themselves are followed over time in the developing animal using mitotrackers or
GFP fusions, they eventually disappear even in autophagy or lysosomal deficient
animals. The same occurs with MOs. In these experiments, it is difficult to rule out
that the structure do not simply loose the marker. The fusion of stabilized mitochon-
dria with the maternal network might, for instance, dilute the marker enough to
prevent their localization.
Furthermore, we are still facing experimental limits. For example, when the
fluorescent markers like GFP::ANT-1 (Al Rawi et al. 2011) or GFP::HSP-6 (Sato
and Sato 2011) were used to follow the sperm mitochondria, they were genetically
encoded and expressed in the germ line by the paternal nuclear genome. This
property rapidly precludes using them for the specific tracking of sperm mitochon-
dria over generations since the marker is transmitted and expressed in the progeny.
So far, these mitochondrial markers were expressed under germ line-specific
promotors (pie-1 or mex-5 promotors) in order to load spermatozoa with the
fluorescent-tagged protein. In these cases, the germ line expression starts again in
the precursor germ cells in late embryogenesis, preventing to follow inherited sperm
mitochondria in these cells after that stage. Regarding the MOs, it is very likely that
they are still degraded in absence of functional autophagy machinery since none of
Autophagosomal Sperm Organelle Clearance and mtDNA Inheritance in C. elegans 21

the genetic background stabilizing the sperm organelles led to the transmission and
accumulation of MO markers over time. For instance, ALLO-1 mutant is viable and
would accumulate MOs over generation if they were stabilized and randomly
segregated within the developing embryo including in the precursor germ cells.
No such accumulation has been observed using an MO-specific antibody.
It is time to address the question of the mechanisms insuring sperm organelle
clearance in a more integrated manner. Several mechanisms are important for normal
sperm organelle degradation, but we still do not have a clear overview of how they
integrate and what is the relative importance of these pathways in the normal
C. elegans embryo. We now have more indications on how the embryo reacts to
sperm organelles entry, but the signal(s) that allows their specific targeting remain
elusive. C. elegans represents an appealing model to conduct unbiased genetic
screens to identify mutants transmitting their sperm mtDNA and such possible
sperm signal. Genetic screens should be conducted in (1) mutated maternal genetic
backgrounds with a delay in sperm organelle degradation phenotype in order to
identify the potential additional mechanisms besides autophagy and (2) on the
paternal side in order to identify the sperm-specific signal for the degradation. The
genetic approach could in principle reveal not only proteins but also posttranslational
modifications or other types of signals like noncoding RNAs. Modern quantitative
and comparative proteomic approaches of sperm and oocytes mitochondria compo-
sition could also reveal a specific factor involved in sperm mitochondria recognition.

Acknowledgments Our laboratory is supported by the Fondation pour la Recherche Medicale

(Equipe FRM DEQ20160334874) and the European COST Program (BM1408 GENiE). We are
grateful to Martin Sachse from Institut Pasteur Paris for the TEM and Charlène Perrois, Valeria
Parrales, and Sebastien Normant for their technical support.

References

Al Rawi S, Louvet-Vallee S, Djeddi A, Sachse M, Culetto E, Hajjar C, Boyd L, Legouis R, Galy V

(2011) Postfertilization autophagy of sperm organelles prevents paternal mitochondrial DNA
transmission. Science 334:1144–1147
Alberti A, Michelet X, Djeddi A, Legouis R (2010) The autophagosomal protein LGG-2 acts
synergistically with LGG-1 in dauer formation and longevity in C. elegans. Autophagy
6:622–633
Anderson S, Bankier AT, Barrell BG, de Bruijn MH, Coulson AR, Drouin J, Eperon IC, Nierlich
DP, Roe BA, Sanger F, Schreier PH, Smith AJ, Staden R, Young IG (1981) Sequence and
organization of the human mitochondrial genome. Nature 290:457–465
Chatterjee I, Richmond A, Putiri E, Shakes DC, Singson A (2005) The Caenorhabditis elegans
spe-38 gene encodes a novel four-pass integral membrane protein required for sperm function at
fertilization. Development 132:2795–2808
Djeddi A, Al Rawi S, Deuve JL, Perrois C, Liu YY, Russeau M, Sachse M, Galy V (2015) Sperm-
inherited organelle clearance in C. elegans relies on LC3-dependent autophagosome targeting to
the pericentrosomal area. Development 142:1705–1716
22 J. Merlet et al.

Eura Y, Ishihara N, Yokota S, Mihara K (2003) Two mitofusin proteins, mammalian homologues of
FZO, with distinct functions are both required for mitochondrial fusion. J Biochem
134:333–344
Fawcett DW (1981) The cell. W. B. Saunders, Philadelphia
Hajjar C, Sampuda KM, Boyd L (2014) Dual roles for ubiquitination in the processing of sperm
organelles after fertilization. BMC Dev Biol 14:6
Italiano JE Jr, Roberts TM, Stewart M, Fontana CA (1996) Reconstitution in vitro of the motile
apparatus from the amoeboid sperm of Ascaris shows that filament assembly and bundling move
membranes. Cell 84:105–114
Jacob WA, Bakker A, Hertsens RC, Biermans W (1994) Mitochondrial matrix granules: their
behavior during changing metabolic situations and their relationship to contact sites between
inner and outer mitochondrial membranes. Microsc Res Tech 27:307–318
Labrousse AM, Zappaterra MD, Rube DA, van der Bliek AM (1999) C. elegans dynamin-related
protein DRP-1 controls severing of the mitochondrial outer membrane. Mol Cell 4(5):815–826
Lambert CC, Epel D (1979) Calcium-mediated mitochondrial movement in ascidian sperm during
fertilization. Dev Biol 69:296–304
Lemmon MA (2008) Membrane recognition by phospholipid-binding domains. Nat Rev Mol Cell
Biol 9:99–111
Liau WS, Gonzalez-Serricchio AS, Deshommes C, Chin K, Lamunyon CW (2007) A persistent
mitochondrial deletion reduces fitness and sperm performance in heteroplasmic populations of
C. elegans. BMC Genet 8:8
Luo SM, Ge ZJ, Wang ZW, Jiang ZZ, Wang ZB, Ouyang YC, Hou Y, Schatten H, Sun QY (2013)
Unique insights into maternal mitochondrial inheritance in mice. Proc Natl Acad Sci U S A
110:13038–13043
Machaca K, L’Hernault SW (1997) The Caenorhabditis elegans spe-5 gene is required for morpho-
genesis of a sperm-specific organelle and is associated with an inherent cold-sensitive pheno-
type. Genetics 146:567–581
Manil-Segalen M, Lefebvre C, Jenzer C, Trichet M, Boulogne C, Satiat-Jeunemaitre B, Legouis R
(2014) The C. elegans LC3 acts downstream of GABARAP to degrade autophagosomes by
interacting with the HOPS subunit VPS39. Dev Cell 28:43–55
Melendez A, Talloczy Z, Seaman M, Eskelinen EL, Hall DH, Levine B (2003) Autophagy genes are
essential for dauer development and life-span extension in C. elegans. Science 301:1387–1391
Nelson GA, Ward S (1980) Vesicle fusion, pseudopod extension and amoeboid motility are induced
in nematode spermatids by the ionophore monensin. Cell 19:457–464
Nishimura H, L’Hernault SW (2017) Spermatogenesis. Curr Biol 27:R988–R994
Nishimura Y, Yoshinari T, Naruse K, Yamada T, Sumi K, Mitani H, Higashiyama T, Kuroiwa T
(2006) Active digestion of sperm mitochondrial DNA in single living sperm revealed by optical
tweezers. Proc Natl Acad Sci U S A 103:1382–1387
Okimoto R, Macfarlane JL, Clary DO, Wolstenholme DR (1992) The mitochondrial genomes of
two nematodes, Caenorhabditis elegans and Ascaris suum. Genetics 130:471–498
Palmisano NJ, Melendez A (2018) Autophagy in C. elegans development. Dev Biol. https://doi.org/
10.1016/j.ydbio.2018.04.009
Politi Y, Gal L, Kalifa Y, Ravid L, Elazar Z, Arama E (2014) Paternal mitochondrial destruction
after fertilization is mediated by a common endocytic and autophagic pathway in Drosophila.
Dev Cell 29:305–320
Roberts TM, Pavalko FM, Ward S (1986) Membrane and cytoplasmic proteins are transported in
the same organelle complex during nematode spermatogenesis. J Cell Biol 102:1787–1796
Rojansky R, Cha MY, Chan DC (2016) Elimination of paternal mitochondria in mouse embryos
occurs through autophagic degradation dependent on PARKIN and MUL1. elife 5:e17896.
https://doi.org/10.7554/eLife.17896
Rolland SG, Lu Y, David CN, Conradt B (2009) The BCL-2-like protein CED-9 of C. elegans
promotes FZO-1/Mfn1,2- and EAT-3/Opa1-dependent mitochondrial fusion. J Cell Biol
186:525–540
Autophagosomal Sperm Organelle Clearance and mtDNA Inheritance in C. elegans 23

Sato M, Sato K (2011) Degradation of paternal mitochondria by fertilization-triggered autophagy in

C. elegans embryos. Science 334:1141–1144
Sato M, Sato K, Tomura K, Kosako H, Sato K (2018) The autophagy receptor ALLO-1 and the
IKKE-1 kinase control clearance of paternal mitochondria in Caenorhabditis elegans. Nat Cell
Biol 20:81–91
Shakes DC, Ward S (1989) Mutations that disrupt the morphogenesis and localization of a sperm-
specific organelle in Caenorhabditis elegans. Dev Biol 134:307–316
Sharpley MS, Marciniak C, Eckel-Mahan K, McManus M, Crimi M, Waymire K, Lin CS,
Masubuchi S, Friend N, Koike M, Chalkia D, Macgregor G, Sassone-Corsi P, Wallace DC
(2012) Heteroplasmy of mouse mtDNA is genetically unstable and results in altered behavior
and cognition. Cell 151:333–343
Smith JR, Stanfield GM (2011) TRY-5 is a sperm-activating protease in Caenorhabditis elegans
seminal fluid. PLoS Genet 7:e1002375
Sutovsky P, Moreno RD, Ramalho-Santos J, Dominko T, Simerly C, Schatten G (1999) Ubiquitin
tag for sperm mitochondria. Nature 402:371–372
Tsang WY, Lemire BD (2002a) Mitochondrial genome content is regulated during nematode
development. Biochem Biophys Res Commun 291:8–16
Tsang WY, Lemire BD (2002b) Stable heteroplasmy but differential inheritance of a large mito-
chondrial DNA deletion in nematodes. Biochem Cell Biol 80:645–654
Tsang WY, Lemire BD (2003) The role of mitochondria in the life of the nematode, Caenorhabditis
elegans. Biochim Biophys Acta 1638:91–105
Tsukamoto S, Kuma A, Murakami M, Kishi C, Yamamoto A, Mizushima N (2008) Autophagy is
essential for preimplantation development of mouse embryos. Science 321:117–120
van der Bliek AM, Shen Q, Kawajiri S (2013) Mechanisms of mitochondrial fission and fusion.
Cold Spring Harb Perspect Biol 5:a011072. https://doi.org/10.1101/cshperspect.a011072
Wang Y, Zhang Y, Chen L, Liang Q, Yin XM, Miao L, Kang BH, Xue D (2016) Kinetics and
specificity of paternal mitochondrial elimination in Caenorhabditis elegans. Nat Commun
7:12569
Ward S, Carrel JS (1979) Fertilization and sperm competition in the nematode Caenorhabditis
elegans. Dev Biol 73:304–321
Ward S, Argon Y, Nelson GA (1981) Sperm morphogenesis in wild-type and fertilization-defective
mutants of Caenorhabditis elegans. J Cell Biol 91:26–44
Washington NL, Ward S (2006) FER-1 regulates Ca2+-mediated membrane fusion during
C. elegans spermatogenesis. J Cell Sci 119:2552–2562
Wei Y, Chiang WC, Sumpter R Jr, Mishra P, Levine B (2017) Prohibitin 2 is an inner mitochondrial
membrane mitophagy receptor. Cell 168:224–238.e10
Wolf N, Hirsh D, Mcintosh JR (1978) Spermatogenesis in males of the free-living nematode,
Caenorhabditis elegans. J Ultrastruct Res 63:155–169
Xu XZ, Sternberg PW (2003) A C. elegans sperm TRP protein required for sperm-egg interactions
during fertilization. Cell 114:285–297
Zamponi N, Zamponi E, Cannas SA, Billoni OV, Helguera PR, Chialvo DR (2018) Mitochondrial
network complexity emerges from fission/fusion dynamics. Sci Rep 8:363
Zhang Y, Yan L, Zhou Z, Yang P, Tian E, Zhang K, Zhao Y, Li Z, Song B, Han J, Miao L, Zhang H
(2009) SEPA-1 mediates the specific recognition and degradation of P granule components by
autophagy in C. elegans. Cell 136:308–321
Zhou Q, Li H, Xue D (2011) Elimination of paternal mitochondria through the lysosomal degra-
dation pathway in C. elegans. Cell Res 21:1662–1669
Zhou Q, Li H, Li H, Nakagawa A, Lin JL, Lee ES, Harry BL, Skeen-Gaar RR, Suehiro Y,
William D, Mitani S, Yuan HS, Kang BH, Xue D (2016) Mitochondrial endonuclease G
mediates breakdown of paternal mitochondria upon fertilization. Science 353:394–399
Doubly Uniparental Inheritance of mtDNA:
An Unappreciated Defiance of a
General Rule

Eleftherios Zouros and George C. Rodakis

Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2 Observations and Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.1 The F and M Content of Somatic and Germ Cell Lines of Mytilus . . . . . . . . . . . . . . . . . . . 27
2.2 The Domination of the Male Germ Line by the M mtDNA Genome . . . . . . . . . . . . . . . . . 30
2.3 The Exclusion of the Maternal mtDNA from the Sperm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.4 Maleness and Presence of Sperm-Transmitted mtDNA: An Associative But Not
Causative Relationship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.5 “Masculinization” or Reversal of the Transmission Route of the F Genome . . . . . . . . . 35
2.6 F/M Phylogeny and the Question of DUI Origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.7 Why Is the M Genome Necessary for Male Fertility? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.8 The One-factor Model for DUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3 Some Outstanding Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

Abstract We recount the basic observations about doubly uniparental inheritance

(DUI) of mtDNA in bivalvian mollusks with an emphasis on those that were
obtained from work in Mytilus and appeared after the review by Zouros (Evol
Biol 40:1–31, 2013). Using this information, we present a new model about DUI
that is a revised version of previously suggested models. The model can be summa-
rized as follows. A Mytilus female either provides its eggs with the “masculinizing”
factor S and the “sperm mitochondria binding” factor Z, or it does not. This property
of the female is determined by two nuclear genes, S and Z, that are always in the
on/on or the off/off phase. In fertilized eggs without factors S and Z the embryo
develops into a female and the sperm mitochondria are randomly dispersed among
cells following development. In fertilized eggs with factors S and Z, the first factor
causes the cell to become eventually sperm and the second causes the sperm

E. Zouros (*)
Department of Biology, University of Crete, Heraklion, Greece
e-mail: [email protected]; [email protected]
G. C. Rodakis
Department of Biology, National and Kapodistrian University of Athens, Athens, Greece

© Springer Nature Switzerland AG 2019 25

P. Sutovsky (ed.), Cellular and Molecular Basis of Mitochondrial Inheritance,
Advances in Anatomy, Embryology and Cell Biology 231,
https://doi.org/10.1007/102_2018_4, Published online: 14 January 2019
26 E. Zouros and G. C. Rodakis

mitochondria to aggregate and anchor to the nuclear membrane by binding to a

specific motif of the sperm-derived mtDNA. Factors S and Z are continuously co-
synthesized and co-localized in the cell line from the egg to the sperm. The sperm
mitochondria of the aggregate escape the mechanism that eliminates the cell’s
mitochondria before the formation of the sperm. The rescued mitochondria are
subsequently packed into five mega-mitochondria in the sperm and are delivered
in the egg.

Keywords Blue mussel, Doubly uniparental inheritance, Mitochondrial DNA

inheritance, Mytilus

1 Introduction

The mitochondrial DNA (mtDNA) was introduced in population genetics, phylog-

eny, and evolution by Avise et al. (1979) and became soon the most powerful tool in
the field (Avise 2004). By that time it was firmly established that animal mtDNA was
maternally inherited (Huthcinson al. 1974; Hayshi et al. 1978), a phenomenon that
became known as strict maternal inheritance (SMI). It was subsequently observed
that the maternal inheritance was not strict. MtDNA heteroplasmy, i.e., the presence
of two or more types of mtDNA in the same individual, was reported in many
animals and in some of these the heteroplasmy was shown to be due to “leakage” of
paternal mtDNA – the erratic escape from a mechanism that prevents the inheritance
of paternal mtDNA (White et al. 2008). Fisher and Skibinski (1990) noted that this
type of heteroplasmy was unusually high in the mussel Mytilus edulis and that male
mussels were more often heteroplasmic than females. They suggested biparental
inheritance as the most probable explanation. Hoeh et al. (1991) also observed a high
level of heteroplasmy for highly diverged mtDNA molecules in Mytilus and
suggested biparental inheritance. Zouros et al. (1992a, b) confirmed biparental
inheritance by analyzing a set of mussel pair-matings. Soon it became obvious that
the mussel heteroplasmy could not be a simple case of paternal mtDNA leakage. A
clear explanation of male heteroplasmy and paternal mtDNA transmission in mus-
sels appeared in two pairs of publications (Skibinski et al. 1994a; Zouros et al.
1994a, b; Skibinski et al. 1994b). These authors described a new system of mtDNA
inheritance, which is known as “doubly uniparental inheritance” (DUI) (Zouros et al.
1994a).
The main features of DUI can be summarized as follows: The species M. edulis
has two clearly distinguishable mtDNA molecules, namely F and M. The F is
transmitted though the egg, and the M through the sperm. Somatic tissues of both
sexes and female gonads are dominated by the F, and male gonads by the M. DUI
was subsequently found in several species of bivalve mollusks, now numbering
more than 100 (Gusman et al. 2016). The basic features described above hold for all
these species, but minor differences may exist. Several reviews of DUI have
appeared, each summarizing the status of our knowledge at the time the review
Doubly Uniparental Inheritance of mtDNA: An Unappreciated Defiance of a. . . 27

was written (Zouros 2000; Breton et al. 2007; Passamonti and Ghiselli 2009; Zouros
2013). Here we give an account of the present status of our understanding of DUI,
paying special attention to results that appeared after the aforementioned reviews.
Most of these results refer to Mytilus edulis – galloprovincialis, a pair of species that
have different yet overlapping geographical distributions, hybridize readily and have
very similar F and M genomes. The reader who wants to get a full account of
observational and experimental data about DUI should, at the minimum, consult the
review of Zouros (2013) and the papers on Mytilus that appeared after this review –
and are cited here. We use the occasion of this review to synthesize whatever we
currently know into a comprehensive model. The strength of empirical evidence
supporting the various elements of the model varies considerably and the likelihood
that some parts of the model, or even the entire model, could be proved wrong
remains high. But it appears to us that this is the best working hypothesis we may
present at this moment.

2 Observations and Models

2.1 The F and M Content of Somatic and Germ Cell Lines

of Mytilus

A key observation that came from mussel pair matings and also from broods of
single females collected from the wild (Kenchington et al. 2002) was that there are
three types of females with regard to the sex ratio of their progeny: females that
produce almost exclusively daughters (daughter biased females), females that pro-
duce sons in high percentages (son-biased females), and females that could not be
assigned statistically to one or the other of the two previous classes (females with
unbiased sex ratio). The origin of sperm that fertilized the eggs did not affect this
property of the female. Given that there are no secondary sexual characteristics in
Mytilus, this classification of females was originally based on the presence or
absence of the M genome in the gonads of their progeny (presence equated to
maleness, absence to femaleness). It was subsequently observed from pedigrees of
laboratory lines that the sex-ratio bias was inherited: daughter-biased mothers
produced more often daughter-biased female progeny than male-biased mothers.
These observations led to conclusion that in Mytilus sex is determined exclusively
by the mother’s nuclear genotype. From previous studies it was known that the
Mytilus sperm contains five (and, rarely, four or six) mitochondria of large size
(Longo and Dornfeld 1967). These “mega-mitochondria” are apparently the product
of fusion of many “normal size” mitochondria (in a way analogous to that known to
occur in the sperm of Drosophila; DeLuca and O’Farrell 2012) and can be easily
seen under the microscope if stained with Mitotracker Green FM.
28 E. Zouros and G. C. Rodakis

Cao et al. (2004a) examined the fate of sperm mitochondria in eggs from
daughter-biased and son-biased mothers. They observed that in eggs that came
from daughter-biased mothers the sperm mitochondria were randomly dispersed in
the egg cytoplasm and in subsequent egg divisions they segregated randomly among
the blastomeres. In eggs that came from son-biased mothers the five sperm mito-
chondria remained in close proximity to each other. In the two-cell stage all five
mitochondria were found in the same blastomere and very close to the cleavage
furrow that separated the two cells. In the four-cell stage the sperm mitochondria
could be seen aligned in the inner part of the cleavage furrow of blastomere D, which
is the blastomere that gives rise to the mesoderm and the germ line (Verdonk and van
den Biggelaar 1983). This type of segregation and positioning close to the cleavage
furrow continued in subsequent cell divisions and in several occasions, when the
sperm mitochondria could still be seen, it was observed to persist as late as the
trochophore stage and even in larvae of stage D, almost 72 h post-fertilization (Cao
et al. 2004a). The two patterns of sperm mitochondria behavior in early embryo-
genesis were named “dispersed” and “aggregate.” The aggregation is not very
faithful. Occasionally, one or even two sperm mitochondria can be found, in
apparently random position, in different blastomeres than the one that contains the
aggregate.
There is nothing special about the distribution of either the F or the M mtDNA
types in female or male somatic tissues (Fig. 1A, Ba). Occasional presence of M in
female somatic tissues is expected from the fact that in eggs destined to develop into
females the sperm mitochondria are not eliminated, but disperse randomly in the
blastomeres. This expectation is consistent with the observation (see Zouros 2013
for references in Mytilus and the clam Ruditapes philippinarum). No regularity was
found in the presence/absence or amount of the M genome among female tissues.
Erratic presence of the M genome in male somatic tissues is also expected from the
occasional break-away of mitochondria from the aggregate that is formed in eggs
destined to become males. Presence of the M genome in male somatic tissues was
observed more often than female somatic tissues, but it was not commoner in tissues
of mesodermic origin, as would be expected if sperm mitochondria broke away from
the aggregate after the formation of the mesoderm.
Of special interest is the question of whether the M genome that leaks into female
embryos may find its way to eggs, as is shown – as a possibility – in Fig. 1. Ghiselli
et al. (2011) failed to detect the M genome in highly purified collections of eggs from
the venerid Ruditapes philippinarum and suggested that there must be a mechanism
of elimination of M during oogenesis. There is no evidence of such a mechanism in
Mytilus. Garrido-Ramos et al. (1998), Obata et al. (2007), and Sano et al. (2007,
2010) reported occasional presence of M in Mytilus eggs along with the dominant
F. Consistent with this is the detection, in low frequency, of triplasmic individuals
that carried one F and two M genomes. Of these, the most abundant was inherited
from the sperm that the other was apparently present in minute amounts in the egg
along with the F (Obata et al. 2007; Theologidis 2007).
Doubly Uniparental Inheritance of mtDNA: An Unappreciated Defiance of a. . . 29

4
A
1 2 3
n
a Female
soma

Zygote type A n

Egg

B
n
a Male
z z soma
z z z
z z z
z z z z b1
z
z zz
zz z
z
Zygote type B z
z n

b2 c
zz z zzzz
zz z z
z z
PGC z n n

b3

Fig. 1 Ontogenesis and gametogenesis in M. edulis. Blue dot: M-carrying mitochondrion, red dot:
F carrying mitochondrion, n: nucleus, right-low cell: blastomere D. (A) Fertilized egg with the
dispersed pattern of sperm mitochondria. The embryo develops into a female. A somatic cell may be
free of M mtDNA or may contain it in tiny amounts. The latter possibility is even less likely for the
egg. (B) Fertilized egg showing the aggregate pattern of sperm mitochondria. The embryo develops
into a male. Z is the hypothesized factor responsible for the formation of the sperm mitochondria
aggregate that binds to the mtDNA of sperm mitochondria and causes them to form an aggregate

The evidence that the sperm is free of the F genome is much stronger. Venetis
et al. (2006) examined for the presence of F sperm of M. galloprovincialis that was
not forced to swim through a solution of percol and sperm that was forced to swim.
They detected the presence of F in 24 of 36 cases in the first and in none in the
second. The study serves also as an illustration that contamination of DNA prepa-
rations from male gonads by the somatic cells containing the F is hard to avoid.
Ghiselli et al. (2011) used also the percol assay and found no F in the sperm of
R. philippinarum.
30 E. Zouros and G. C. Rodakis

2.2 The Domination of the Male Germ Line by the M mtDNA

Genome

We have no data bearing on the question of how the M genome becomes the
dominant mtDNA in the male gonad from being a minority in the fertilized egg,
beyond the observation that in Mytilus the sperm mitochondria aggregate remains
until stage D (Cao et al. 2004a). Thus, we may only list alternative hypotheses
(Fig. 1). One hypothesis is that the route of the sperm mitochondria that we described
for the early egg divisions continues until the formation of the first primordial germ
cell (PGC). We may call it “the PGC delivery” hypothesis to emphasize that there is
a mechanism in the male embryo that directs and delivers the sperm mitochondria
aggregate in the PGC. The mechanism does not prevent detachment or loss of sperm
mitochondria along the way. But it secures that all or some of these mitochondria
will end in the PGC rather than in somatic tissues or be lost entirely. In Fig. 1 the
PGC delivery hypothesis is presented by cases b1 and b2. There is an important
difference between these two cases. In case b1 it is assumed that there is a mecha-
nism that allows only the aggregate to enter the PGC. In case b2 there is no exclusion
of the F mtDNA from the PGC. Case b3 is the alternative of the PGC delivery
hypothesis. According to it the sperm mitochondria aggregate is dissolved before the
formation of the PGC and sperm mitochondria divide along with egg mitochondria.
The three hypotheses about the formation of the PGC lead to different hypotheses
about the presence of the F and M genomes in the male gonad. In case b1 the
exclusion of the F genome from the PGC explains at once why the male gonad is
dominated by the M genome and why the sperm contains only this genome. But case
b1 requires an explanation about the mechanism of delivery of the sperm mitochon-
dria aggregate in the PGC and the exclusion of the F genome from the PGC. Case b2
also needs an explanation of how the sperm mitochondria are delivered in the PGC,
but it also requires an explanation of how the M becomes dominant in the gonad in
the presence of the F genome and how the F genome is eliminated from the mature
sperm. The last two questions apply also to case b3.
All versions that we have listed above have been mentioned as possibilities in the
studies that dealt with the presence of the F genome in the male germ line (see Zouros
2013 for references). These studies produced unclear and, sometimes, conflicting
results, which, in large degree, can be attributed to difficulty to eliminate contamina-
tion of mtDNA preparations from male gonads with somatic cells that are rich in F
mtDNA. The quantitative study of Obata et al. (2011) of the amount and the degree of
expression of the F and the M genomes in male gonads of M. galloprovincialis
produced a picture of widely varying F/M ratios in various stages of male gonad
development. Interestingly, these authors observed expression of the F type in
spermatogonia and spermatocytes but not in spermatids and spermatozoa. At present
the consensus is that the F genome is present in the male germ line and that its
elimination from the sperm occurs at some late stage of sperm formation.
Doubly Uniparental Inheritance of mtDNA: An Unappreciated Defiance of a. . . 31

2.3 The Exclusion of the Maternal mtDNA from the Sperm

The elimination of mtDNA from the sperm is the rule in the animal kingdom.
Assuming that the F genome corresponds to the “standard” type on animal
mtDNA, its elimination from the sperm is the expected outcome. Sutovsky et al.
(1999, 2000) have shown that the mechanism of mtDNA elimination from sperm in
bovines and monkeys involves ubiquitination of the membrane of sperm mitochon-
dria and their subsequent disposal. It is, therefore, reasonable to depart from the
assumption that an analogous mechanism exists in Mytilus (and, by extension, to all
species with DUI) and that the M genome is immune to this mechanism. As we have
noted, there is good evidence in the literature for the presence of the F genome in the
male germ line. If so, how its elimination from the mature sperm is achieved? One
possibility is that the F or the M genome, or both, code for an RNA or a protein factor
which is an indispensible part of the mechanism. The other possibility is that the
mechanism involves a cytoplasmic factor which somehow recognizes and selec-
tively destroys the F mtDNA carrying mitochondria or, inversely, recognizes and
protects the M mtDNA carrying ones.
The existence of open reading frames (ORF) in both mtDNA types in Mytilus and
in unionids favored the hypothesis of an mtDNA coded factor. The hypothesis got
further support from the detection of a protein that corresponds to an ORF in the F
genome of fresh water mussels (Breton et al. 2009, 2011a). The association of this
protein with exclusion of the F genome from the sperm in unionids remains to be
established. The possibility that F or M ORFs code for a protein or an RNA product
in Mytilus is remote. Chatzoglou et al. (2013) studied the mRNAs coded by both the
F and M genomes of M. galloprovincialis. They identified the mRNAs of all mtDNA
coded proteins in the primary transcript and the points of its cleavage in eight
monocistronic, one tricistronic, and one possible bicistronic product. But they failed
to identify any mRNA sequence that could match an ORF in either the F or the M
mtDNA. A more detailed study of the transcriptosome of the same species by
Kyriakou et al. (2014a, b) also produced no evidence for a product coded by an
ORF of the M or the F genome.
The shift of attention to the possibility that the mtDNAs of Mytilus may contain
sequences that interact with a cytoplasmic factor (or factors) produced more encour-
aging results. Kyriakou et al. (2015) performed a series of electrophoretic mobility
shift assays (EMSA) with extracts from male or female gonad as the protein-factor
source and with DNA clones spanning the control region (CR) of the M and F
genomes as binding factors. The control region (CR) of both genomes has been
divided into three domains on the basis of primary sequence variability among F,
among M or divergence between M and F genomes (Fig. 2). The first domain (in the
50 –30 direction) is highly variable (thus called first variable domain, VD1), the
second is highly conserved (conserved domain, CD), and the third is a short highly
variable piece (VD2). Kyriakou et al. (2015) identified a sequence of 151 bp in the
32 E. Zouros and G. C. Rodakis

control region (CR)

Part of VD1 that exists
in C genome

835
902
17671 1 151 472 473 834
17623

M-type l-rRNA VD1 CD VD2 tRNATyr

VD1-M161-312 161 312 CD-M592-796 592 796

VD1-M148-219 148 219
VD1-M231-312 231 312
VD1-M233-255 233 255
VD1-M180-255 180 255

100 bp

Fig. 2 Identification of a sequence motif in the M genome that interacts with a cytoplasmic factor.
The main control region (CR) of the F and the M mtDNA of species of the genus Mytilus are
flanked by the large ribosomal RNA (lrRNA) and the transfer RNA of tyrosine (tRNAtyr). The lower
part shows the clones that were used in EMSA assays. Dark box: the region that binds to protein
factor; grey box: the polyA track that is necessary to bend the mtDNA for the facilitation of the
binding; white box: a helper part of no-specific sequence. The check mark identifies the clones that
produced the complex with proteins from the perinuclear cytoplasm. The smallest tested length of
M mtDNA with the ability to produce the complex was 75 bp

VD1 region of the M genome that produced a shift with the extracts from the male
gonad. Within this 151 bp segment there is a highly conserved element of 23 bp
which is preceded by a poly-A sequence. Both the 23 bp element and the poly-A
track are indispensable for the formation of the complex. The stretch upstream of the
poly-A track is also needed for the formation of the complex, but need not have a
specific sequence. Further, the authors experimented separately with nuclear and
cytoplasmic extracts of the male gonad and found that the shift occurred with the
nuclear but not the cytoplasmic extract. The shift was not produced when they used
the same clones with extracts from eggs or from somatic tissues of males or females.
Clones from the VD1 of the F genome produced no shift with extracts from male or
female gonads. As a kind of control for the EMSA assay, Kyriakou et al. (2013) had
previously used clones from the CD of the F and the M mtDNA. The CD contains
sequences that elicit the molecule’s replication and transmission. As expected, there
was a shift with protein extracts from the cytoplasm.
The model proposed by Kyriakou et al. (2015) for the elimination of the F
genome from the mature sperm is shown in Fig. 3. A cell that is destined to produce
a spermatozoon is assumed to contain M-carrying and F-carrying mitochondria and a
“protection factor,” Z, which is located close to the nuclear membrane. Because the
authors used for protein extraction whole male gonads (which consist of cells at
various stages of maturation), cell 1 of Fig. 3 can, theoretically, be of any stage of
Doubly Uniparental Inheritance of mtDNA: An Unappreciated Defiance of a. . . 33

4
1 2 3

zz z zz
zz z
z z
n n n

Fig. 3 A model for the exclusion of F mtDNA from the sperm. Blue dot: M-carrying mitochon-
drion, red dot: F carrying mitochondrion, n: nucleus, Z the hypothetical “protector” factor. Stage 1:
Male gonadal cell. Stage 2: Gonadal cell just before it becomes spermatid. Stage 3: Spermatid.
Stage 4: Mature sperm. Mitochondria marked with x denote mitochondria that are destroyed in late
spermatogenesis by a mechanism that is assumed to be common to species with and without DUI. Z
is the same factor shown in Fig. 1

maturation, from early descendants of PGC to just before spermatid formation. At

stage 2, factor Z binds to M-carrying mitochondria that are around the nucleus.
Again, the binding of factor Z with M-carrying mitochondria and the positional
association of the complex with the nuclear membrane may have occurred at any
stage in the development of the male gonad. At stage 2, a cytoplasmic factor (not
shown) attacks and destroys the mitochondria irrespective of their DNA content,
except those that have formed a complex with factor Z. F-carrying mitochondria are
unprotected and destroyed and so are M-carrying mitochondria located away from
the nucleus. The result is that spermatids (Fig. 3) and mature sperm (Figs. 3 and 4)
contain only the M genome. This is in agreement with the results of Obata et al.
(2011), who observed expression of the F type in spermatogonia and spermatocytes
but not in spermatids and spermatozoa.
The model assumes that the mechanism of sperm mitochondria destruction
(which is, apparently, present in all animals in one form or another) in Mytilus
(and perhaps in all species with DUI, if not all bivalvian mollusks) acts at the
pre-spermatid stage. There are two possibilities worth of consideration for the
transition from the spermatid to the spermatozoon. The first is that the aggregate
pattern of the five mega-mitochondria that is seen in the spermatozoon is formed
concurrently with or immediately after the destruction of the non-protected mito-
chondria. The second is that rescued mitochondria divide and produce a large
number of mitochondria from which the aggregate is formed along with the forma-
tion of the mature sperm.
34 E. Zouros and G. C. Rodakis

parents
gametes
sex genotype sex type frequency
male ee [R], Ee [H], EE [D] sperm (Z–S0) [1]
– 1
female ee [r] eggs (Z S ) [1]

female eE [h] eggs (Z–S1) [k]

(Z+S2) [1–k]

female EE [d] eggs (Z+S2) [1]

offspring
sperm eggs
(Z–S0) (Z–S1) (Z+S2)

sex: female sex: male

mitochondria pattern: dispersed mitochondrial pattern: aggregate

soma: maternal mtDNA soma: maternal mtDNA

gametes: maternal mtDNA gametes: paternal mtDNA

population

h = [(1–(1–2k(1–k))1/2)/(2k(1–k))] H = [(1–4d2)/2]

d = [(1–2h(1–k))/2] D = [(1+2d)2/4]

r = [(1–2hk)/2] R = [(1–2d)2/4]

Fig. 4 The genetic model for the explanation of the major features of DUI. The upper part gives the
types of gametes produced by males and females of a different genotype at locus E. R, H, and D are
the population frequencies of the three genotypes among males, and r, h, and d among females.
Diamond is locus Z and squares are copies of locus S. Dark denotes unconditional inactivation, gray
inactive state, and white active state. Superscripts of S denote the number of active copies of S, Z
and Z+ denote that locus Z is expressed or not expressed. The middle part gives the “phenotypes” of
progeny resulting from any combination of sperm and egg. The lower part gives the equilibrium
frequencies of the three genotypes at locus E as a function of k

2.4 Maleness and Presence of Sperm-Transmitted mtDNA:

An Associative But Not Causative Relationship

As we have mentioned, sex determination in Mytilus is under the control of the

female’s nuclear genes and there is no role left for the male parent. Kiyomoto et al.
(1996) induced triploidy in M. galloprovincialis and observed that all resulting
individuals developed male gonads. Kenchington et al. (2009) treated eggs of
M. edulis from mothers that produced almost exclusively daughters with
cytochalasin B, which arrests meiosis at the first or second division, and examined
whether (a) the sperm mitochondria aggregate was formed in fertilized eggs, (b) the
resulting progeny developed a male or female gonad, and (c) the M genome was
present in the gonad. They also examined no cytochalasin treated eggs from the same
Doubly Uniparental Inheritance of mtDNA: An Unappreciated Defiance of a. . . 35

crosses for the same three characteristics. They observed that in treated eggs there
was no aggregate formation, adult offspring from these eggs produced male gonads,
and these gonads did not contain the M genome. Eggs from the same crosses that
were not treated with cytochalasin B did not show the sperm aggregate and devel-
oped female gonads that lacked the M genome. They repeated the experiment with
eggs derived from mothers that produced mainly sons. The majority of eggs showed
the aggregate sperm in both treated and untreated eggs and developed male gonads
that contained the M genome. The only difference between male gonads from
untreated and treated eggs was that the gonad was normal in the first and with
irregularities of different kinds in the second.
These results point directly to the following conclusions. Formation of the sperm
aggregate and presence of the paternal mtDNA genome in the gonad of the resulting
individual, and, reversely, lack of the aggregate and lack of the paternal genome, go
together and are determined by the mother’s nuclear genotype. Maleness is also
determined by the mother’s nuclear genotype, but it does not require formation of the
sperm mitochondria aggregate or the presence of the paternal mtDNA genome in the
gonad. Maleness is not dependent on the presence of the paternal mtDNA genome.
Under normal circumstances (i.e., when the cross is homospecific and the eggs are
not diplodized) maleness and sperm mtDNA co-occur. The simplest assumption of
this coupling would be that the nuclear factor which is expressed in the female parent
and causes maleness in her offspring and the factor that allows the sperm mtDNA to
enter the germ line of these offspring are tightly linked in the nuclear genome and
co-expressed in oogenesis. Triploidization interferes with sex determination but not
with the fate of sperm mtDNA.

2.5 “Masculinization” or Reversal of the Transmission Route

of the F Genome

The key observation that led to the discovery of DUI was the presence in any mussel
population of two highly divergent mtDNA molecules of which one was transmitted
though the egg and the other through the sperm. This picture was clouded by the
detection of mtDNA molecules whose primary sequence highly resembled the
sequence of the F genome, yet they were transmitted by the sperm. At the beginning
these genomes were considered as F genomes which had “invaded” the male-
transmitted mtDNA lineage and were, since then, transmitted by the sperm. They
were also called “masculinized” genomes, in anticipation of “feminized” genomes,
i.e., M genomes that might have invaded the female-transmitted lineage.
No feminized genomes were observed to this date. Masculinization presented,
however, a major challenge for the understanding of DUI. Happily the problem was
dissolved in a way that turned masculinization into a tool for the understanding of the
molecular mechanism of DUI and its phylogenetic distribution in bivalve mollusks.
One consistent observation in Mytilus edulis/galloprovincialis and the closely
36 E. Zouros and G. C. Rodakis

related species M. trossulus was that the masculinized genomes differed from the
“true” (egg-transmitted) F genomes in the VD1 region of the control region (CR),
thus making this region the primary suspect for sequences involved in the determi-
nation of the route of transmission (Burzynski et al. 2003; Cao et al. 2004b). Soon
afterword it was shown that recombination between mtDNA molecules is possible in
mussels (Ladoukakis and Zouros 2001) and that it had occurred in the history of
several animal species (Piganeau et al. 2004; Tsaousis et al. 2005). Thus, the
hypothesis that “masculinizing” sequences were transferred from the M genome to
the F via recombination became very appealing. The conserved domain (CD) of the
control region maintains the highest sequence similarity between the F and M
genomes in Mytilus and could act as the site of the recombination.
The hypothesis was tested by obtaining the full sequence of the so-called C
genome (Venetis et al. 2007). This is a masculinized genome that occurs in low
frequencies in M. galloprovincialis populations, but it may reach high frequencies in
some populations (Ladoukakis et al. 2002). Venetis et al. (2007) found that the
protein and RNA coding parts of genome C differed little from the standard
egg-transmitted F genome, but the CR was very different (Fig. 5). It consisted of
four tandem repeats of the typical CR. The first and the fourth copy were mosaics of
F and M sequences but the second and third were identical to the control region
of the M genome with the exception of the first 199 bp of the VD1 domain that were
missing. The most parsimonious explanation of the emergence of genome C is a
duplication of the CR of the M genome which allowed for a misalignment of the M
and F genomes and for a crossing over at the CD that shares high sequence
similarities in the two genomes. Interestingly the compound CR of the C genome
contains three identical sequences each of which contains the full 151 bp that
Kyriakou et al. (2015) identified as the sequence necessary for the binding of the
M genome with a factor in the perinuclear extract of male gonads (Fig. 2). The
inference is that the C genome is rescued from the mitochondrial destruction that
occurs in the pre-spermatid stage in the same way as the standard M genome.

2.6 F/M Phylogeny and the Question of DUI Origin

Figure 6 shows the phylogeny of 24 mtDNA genomes (13 F and 11 M) from

fourteen DUI species with very different phylogenetic relationships. Several points
of interest emerge. Within Mytiloida gender-specific clustering was observed only
for the closely related species M. edulis/M. trossulus (which are clearly different
species even though they may occasionally hybridize). The M genome of the
cogeneric species M. californianus clustered away from its conspecific F and the
four genomes of the other two species, reflecting the fast evolution of the M lineage.
The two genomes of the other three mytilid species, G. demissa, B. exustus, and
M. senhusia, clustered according to the origin of species. The same occurred with the
venerid V. philippinarum. In contrast six F and four M genomes form six unionid
Doubly Uniparental Inheritance of mtDNA: An Unappreciated Defiance of a. . . 37

CR

l-rRNA VD1 CD VD2 Y Cyt b M-type

CR

l-rRNA VD1 CD VD2 Y Cyt b F-type

CR-1 CR-2 CR-3 CR-4

l-rRNA VD1.1 CD.1 VD2 ΔVD1.2 CD.2 VD2 ΔVD1.3 CD.3 VD2 ΔVD1.4 CD.4 VD2.4 Y Cyt b
.1 .2 .3
YΔ.1 YΔ.2 YΔ.3

CR Y
D
IQ
l-r

tb
RN

C
E Cy
A

N K
II M2
G s
-rR CO
NA L1
F C-mtDNA L2
ND6 ND1
19109 bp
V
ND5 ND
4

4L CO
ND III
A6

ND

T
COI

S2
ND3

R M1
P W
H A
S1

Fig. 5 The paternally inherited C mtDNA of Mytilus galloprovincialis. The entire genome is
shown in circular form. Letters are the standard symbols for protein and RNA genes. Black boxes
are replication and transcription control regions. The DNA sequence of all coding genes and of the
two minor control regions are F-type. The CR of the C genome consists of three repeats of the CR of
the M genome inserted between the VD1 and the Y gene of the F genome. Δ stands for deletion

species, some of which are as distantly related as the mytilid genera, formed an
impressive gender-specific phylogeny.
A model that explains these observations is shown in Fig. 7. Part A shows three
pairs of species of different phylogenetic relationship. It is assumed that the F and M
genomes diverged from one common mtDNA in the distant past. Species 1 and
2 have recently diverged from a common ancestor species from which they inherited
the same F and the same M genome. In time the F genomes that were “trapped” in
different species diverged from each other and so did the M genomes, the latter at a
higher rate. If scored at this stage the four genomes would cluster according to their
mode of inheritance rather than taxon origin (gender-joining phylogeny). Given
enough time a new sperm-transmitted genome, M3, will emerge from the F genome
(first masculinization event) and may eventually replace M1. At the moment of
masculinization the F/M divergence will be set to zero (the two genomes will differ
only at the CR where the recombination that transferred the masculinizing sequences
from the M to the F occurred). The descendant species 3 will contain this masculin-
ized genome along with the F of species 1, which will have diverged and recognized
38 E. Zouros and G. C. Rodakis

Pectinoida
Ostreoida
Geukensia demissa M
Geukensia demissa F
Brachidontes exustus M

Pteriomorphia
Brachidontes exustus F
Mytiloida
Musculista senhusia M
Musculista senhusia F
Mytilus californianus M
Mytilus californianus F
Mytilus trossulus M
Mytilus edulis M
Mytilus trossulus F
Mytilus edulis F

Veneroida Venerupis philippinarum M

Venerupis philippinarum F
Pyganodon grandis M
Inversidens japanensis M

Heterodonta
Quadrula quadrula M
Venustaconcha ellipsiformis M
Unionida Lampsilis ornata F
Venustaconcha ellipsiformis F
Quadrula quadrula F
Hyriopsis cumingii F
Inversidens japanensis F
Pyganodon grandis F
Octopus vulgaris

Fig. 6 The phylogenetic relationship of maternally (F) and paternally (M) transmitted mtDNA
genomes of species from three super-families of bivalvian mollusks. Phylogenies involving a much
larger number of species (Gusman et al. 2016) are consistent with the conclusions drawn from the
above set of species (taken from Zouros 2013; original data from Theologidis et al. 2008 and
Doucet-Beaupre et al. 2010)

as different (F3). If the four genomes of species 3 and 4 are grouped they will
produce a tree that will reflect neither the mode of inheritance nor the taxon origin
(mixed phylogeny). For two distantly related species (species 5 and 6) the time of
divergence from the common ancestor species would be long enough for a mascu-
linization to occur and replace the ancestral M. If this happened, the comparison of
Doubly Uniparental Inheritance of mtDNA: An Unappreciated Defiance of a. . . 39

distantly
A related
taxa
related
taxa
closely
related M3 M5
taxa
ancestral F1
species

F3 F5

M1
X time
M6

F2 F4

F6

M2 M4
X

B M1 M2 M3 M4 M5 M6
F1 F2 F3 F4 F5 F6

gender-joining mixed taxon-joining

phylogeny phylogeny phylogeny

Fig. 7 How masculinization may explain the various phylogenetic patterns among F and M
genomes from different taxa. (a) Three pairs of species of different phylogenetic distance. Dots
denote masculinization events and X extinction events. (b) The three types of clustering of F and M
genomes from a pair of species

species 5 and 6 would produce a taxon-joining phylogeny, i.e., one in which the two
genomes of each species cluster together.
Figure 6 implies that multiple masculinization events followed by replacement of
the old M have occurred in the long time (perhaps hundreds of millions years) that
separated the deep branches of bivalvian mollusks, such as Pteriomorpha and
Heterodonta. In Mytiloida we have all three types of phylogeny. There has been
no fixed masculinization during the time that separated M. edulis from M. trossulus.
One such event occurred after the lineage that produced M. californianus broke
away from the lineage that produced the two previous species. All four mytiloid
genera have been separated for a time long enough for multiple masculinization
events to occur in the lineages that produced them. The simplest explanation for the
gender-joining phylogeny of the unionid species is that in these species there can be
no crossing over between the M and F genomes of the type that would transfer
masculinizing sequences from the first to the second. This may be either because the
conserved domains of the F and M phylads (the analogs of the CD of Mytilus) have
diverged to the point that no recombination may occur or because the masculinizing
sequences are not in the vicinity of conserved sequences.
The presence/absence of DUI in bivalve mollusks is spotty and irregular
(Theologidis et al. 2008). On face value, this suggests either that DUI emerged
once in the phylum of molluscan bivalves and was subsequently lost in several
lineages or arose independently in several lineages. The evidence we have at present
is not conclusive on this point. If DUI arose only once, the event must be very old,
perhaps as old as 400 million years. If DUI has multiple and independent origins, its
molecular mechanism must be fairly simple and only its details may vary among
40 E. Zouros and G. C. Rodakis

distant species that share the phenomenon. Such differences have, indeed, been
observed. One is the aforementioned lack of masculinization in unionids and the
amount of the M genome in somatic tissues. But, again, such variations may be
expected under the hypothesis of single origin and the very long time that separates
the major phylogenetic divisions of the super-class Autolamellibranchia that con-
tains all known species with DUI. An answer could be obtained if the presence of
DUI was found to co-occur consistently with some other basic feature of the species’
biology. Breton et al. (2011b) observed discordance between DUI and hermaphro-
ditism in a small collection of unionid species in which hermaphroditism appeared to
have arisen independently. They suggested that the transition from dioeciousness to
hermaphroditism is accompanied with DUI loss. The idea that hermaphroditism is
incompatible with DUI is interesting and must be looked in more species with
different taxonomic relationships.
The one-factor model that we describe here (see Sect. 2.8) has a strong bearing on
the question of the origin of DUI. In a species that has the sex-determining mechanism
that we proposed for Mytilus all that is required for the appearance of DUI is the
emergence of a gene, Z, in close proximity to the on/off copy of the sex determining
locus S (Fig. 4). This would mean that a species in which females are either daughter
or son producers is vulnerable to becoming a DUI species. The inverse does not
follow. Whereas any species with DUI will have females with highly biased sex-ratio
progeny, not all species with this feature will have DUI. Under the one-factor model
the multiple origin of DUI may not be highly improbable and two or more species with
DUI need not be similar in all details. Loss of DUI is more problematic. If high
sex-ratio biases of opposite direction among females is a condition for the operation of
DUI, then loss of the bias in a species, for whatever reason, would mean the collapse of
DUI and the transition to standard uniparental transmission of maternal mtDNA. Also,
high sex ratio bias would elevate founder and small population size effects, making, in
the long run, DUI species more vulnerable to extinction compared to species with
standard maternal mtDNA inheritance. This might be one way to explain the presence
of a species with standard maternal inheritance in a cluster of species or families that
contain mostly DUI species.

2.7 Why Is the M Genome Necessary for Male Fertility?

As we have argued in a previous section, a specific short track of noncoding

sequences from the M genome is capable to turn an F genome from female-
transmitted to male-transmitted. Far from vanishing, these newly masculinized F
genomes may replace eventually the old M from the entire species. This suggests
that there is nothing the M genome does that the F genome cannot do, except to ride
with the sperm. The short DNA sequence that the F genome “extracts” from the M is
enough for the newly masculinized F to make the donor M superfluous for the
species. But just riding by the sperm cannot be the only thing the M genome does. If
that was the case, then an “old” M and a “new” M ought to be equal in their
performance as sperm riders, and there would be no explanation of the replacement
Other documents randomly have
different content
 The Dureta, or Reclining Couch, used in the Bath; both in the Calidarium and Frigidarium. (Pages 53, 97.)
a. The plane for supporting the back. b. The thigh-plane. c. The leg-plane. d. The foot-piece, which is
movable, and admits of adjustment, to suit the comfort of the bather. e. The head-piece, for supporting
the head. f. Arc of the angle a-b. g. The angle corresponding with the bend of the knee. h. Arc of the
angle b-c. i. Lower hole, for the foot-piece. k. Elevation of the trunk-plane from the ground-line, l. m.
One of the feet of the couch.
This figure is intended to exhibit the construction of the dureta, the best lines of angle, and the size the
most convenient for a person of medium stature—say five feet, eight inches. If the person be taller or
shorter, a corresponding difference must be made in the length of the three principal pieces. The dureta
is constructed of deal boards, 20 inches long, nailed on a pair of lateral rails; the rails being supported by
a firm foot, m, and steadied by a bracket at the angle g. The measurements are as follow:—a, 28 inches;
b, 18-1/2 inches; c, from g to the foot-piece, 19 inches; and from g to the extreme end, below the letter
i, 23 inches. The holes for the foot-piece are two inches apart; and the head-piece may be made
movable. The arc of the angle a-b, measured at f, from the upper angle of a to g is 38 inches; the arc of
the angle b-c, measured at h from the angle m to the end of the plane c, is 37 inches; the height of the
upper end at k is 24-1/2 inches; the height of the angle g at l is 14-1/2 inches; and the dotted line from
the angle m to the perpendicular k, 21 inches; the height of the angle m from the point where the dotted
line touches to the ground is 6-1/4 inches; and the height of the end at i is simply the depth of the rail—
namely between two and three inches.
The dureta after the above model is manufactured by Mr. Allen, 7, Great Smith-street, Westminster.
 THE

EASTERN, OR TURKISH BATH.

 CHAPTER I.
The Bath is an animal instinct: and, par excellence, a human
instinct; it is as much a necessity of our nature as drink. We drink
because we thirst—an interior sense. We bathe because water, the
material of drink, is a desire of the outward man—an exterior sense.
An animal, whether beast or bird, pasturing or straying near a limpid
stream, first satisfies the inward sense, and then delights the
outward sense. A man, be he savage or civilized, can no more resist
the gratification of bathing his wearied limbs in a warm transparent
pool than he can resist the cup of water when athirst. Instinct bids
him bathe and be clean. To inquire—Who invented the act of
drinking? would be as reasonable as to ask—Who invented the bath?
The bath is coeval with the earliest existence of man. Can it be
doubted that our first parents bathed their newly-created limbs in
the river that "went out of Eden to water the garden"? History
teaches us, that the Phœnicians and ancient Greeks of all ranks,
from the daughters of their kings down to the poorest citizens, were
wont to bathe in rivers and in the sea, for the purpose of cleansing
their bodies and refreshing and invigorating their frames. They had
recourse to the bath when they ceased from sorrow and mourning,
after great fatigues of whatever kind, before and during their meals,
and at the conclusion of their battles. Bathing was the first act of
their lives, and it was a part of their funereal rites. The birth of
Jupiter, the Thunderer, is celebrated by the poet Callimachus in the
following lines:—
"As soon as you were born and saw the light,
Your mother's grateful burden and delight,
She sought for some clear brook to purify
The body of so dear a progeny."
Again, of Alcestis, when about to lay down her life for her husband
Admetus, it is written:—
 "The pious dame, before the fatal day
Of her own exit, bathed her beauteous limbs
In gentle rivulet."
Plato, also, records how the good old philosopher Socrates, before
he drank the fatal cup of hemlock that was to consign him to Hades,
bathed and washed himself, that he might save the women, whose
duty it was, their troublesome office.[1]
A short stage in the history of the bath leads us to the discovery of
springs of hot water, hot vapour, and hot air; and these very possibly
suggested to man's inventive mind the means of procuring so great
a luxury by his own contrivance. Homer commends one of the
sources of the Scamander for its warmth, and tells us how
Andromache, with matronly care, prepared a hot bath for her
husband Hector, against his return from battle:—
"Her fair-haired handmaids heat the brazen urn,
The Bath preparing for her lord's return."
We are taught also that Vulcan, or, as others say, Minerva,
discovered certain hot baths ([Greek: Hêrakleea loutra]) to Hercules,
that he might replenish his strength after undergoing severe exertion
and fatigue. And the Phœdrians, according to Homer, laid great
stress upon the importance to the health and happiness of man of
frequent changes of apparel, comfortable beds, and hot baths.
It is one of the marvels of the earth's history, that hot springs, or
thermal springs, bubble upwards to the light, not only on Mount Ida,
the source of the Scamander, but in countless other places and
countries on the world's surface. These hot springs would appear to
have invited man to their use by their pleasant aspect and by their
warmth; and their enjoyment to have suggested the possibility of
contriving artificially a similar luxury nearer to his threshold.
The word Hamâm, which is equivalent to thermal springs, is not
unfrequently met with in the East as the name of a town or village in
or near to which hot springs are found. Hamâm Ali, in the
neighbourhood of ancient Nineveh, is an example of this kind. "The
thermal spring is covered by a building, only commodious for half-
savage people, yet the place is much frequented by persons of the
better classes both from Baghdad and Mósul."[2] Captain Kennedy, in
his "Travels in Algeria and Tunis," tells us of the hot springs of
Hamâm Meskhoutin, which rise to the surface at a temperature of
203° of Fahrenheit, only 9° short of boiling, and are so abundant as
to burst forth through any opening made accidentally in the ground.
"The thermal waters, in flowing over the bank of the rivulet, have
formed a calcareous deposit of great beauty, resembling a cascade
of the purest white marble, tinged here and there with various
shades of green and orange."
In Italy, near the town of Pozzuoli, are some natural thermal springs
—the ancient Posidianæ, now called the Baths of Nero, of which the
temperature of the water is 185°, while that of the vapour which
rises from it is 122°. The spring is situated in a rocky cavern at the
end of a long passage formed by a fissure in the rock, and in this
way constitutes a natural bathing house.
In Germany, among others, are the thermal springs of Borcette, with
a temperature of 171°; Carlsbad, in Bohemia, 165°; Wiesbaden,
Ems, and Schlangenbad, in Nassau; Baden-Baden; Aix-la-Chapelle;
Wildbad; and Ischl.
In Iceland are the far-famed Geysers; in the Southern Ocean the hot
springs of Amsterdam Island; and many more are dispersed over the
Continent of America; while in England there are the thermal springs
of Bath, Bristol, Buxton, and Matlock.
The heated rock and the vaporization of water would seem to have
originated the primitive idea of a hot-air and hot-vapour bath; and
this idea we find carried out simultaneously in various parts of the
world and amongst the rudest nations. Mr. Gent, in his "History of
Virginia," describes the hot-vapour bath as employed by the
American Indians.
"The doctor," he says, "takes three or four large stones, which, after
having heated red-hot, he places in the middle of the stove, laying
on them some of the inner bark of oak, beaten in a mortar, to keep
them from burning; this being done, they (the Indians) creep in, six
or eight at a time, or as many as the place will hold, and then close
up the mouth of the stove, which is usually made like an oven in
some bank near the water-side; in the meanwhile, the doctor, to
raise a steam, after they have been stewing a little time, pours cold
water on the stones, and now and then sprinkles the men to keep
them from fainting; after they have sweat as long as they can well
endure it, they sally out, and (though it be in the depth of winter)
forthwith plunge themselves over head and ears in cold water, which
instantly closes up the pores and preserves them from taking cold."
After the bath, they are anointed like the Romans, the pomatum of
the Indians being for the most part bear's-grease, containing a
powder obtained by grinding the root of the yellow alkanet.
But we find this primitive form of bath nearer home than the
American Continent—namely, in Ireland, although both the American
and the Irish bath may, Mr. Urquhart suggests, have been derived
from the same ancestry—that of the Phœnicians. In a foot-note
appended to a page on the universality of the bath, in his "Pillars of
Hercules,"[3] Mr. Urquhart gives the following very curious and very
interesting account of the practice of sweating employed in former
times in Ireland, as reported to him by a lady as a recollection of her
childhood:—
"With respect to the sweating-houses, as they are called, I
remember about forty years ago seeing one in the island of Rathlin,
and shall try to give you a description of it. It was built of basalt
stones, very much in the shape of a bee-hive, with a row of stones
inside, for the person to sit on when undergoing the operation.
There was a hole at the top, and one near the ground, where the
person crept in and seated him or herself, the stones having been
heated in the same way as an oven for baking bread is, the hole on
the top being covered with a sod while being heated, but I suppose
removed to admit the person to breathe. Before entering, the
patient was stripped quite naked, and on coming out, dressed again
in the open air. The process was reckoned a sovereign cure for
rheumatism and all sorts of pains and aches."
Dr. Haughton on the same subject remarks that:—"Two varieties of
Tig Allui, or sweating-houses, exist in Ireland, one kind being
capable of containing a good many persons, and the other only
intended for a single occupant." The former is that just described: it
is heated in the same way as an oven, by making a large fire of
wood in the middle of the floor, and after the wood is burnt out,
sweeping away the ashes. Besides the cure of rheumatism, the
young girls who have tarnished their complexion in the process of
burning kelp or sea-weed for the manufacture of soda, also resort to
the Tig Allui for the purpose of clearing their skin. The usual time for
remaining in the bath is under the half hour.
The second kind of Tig Allui—namely, that for the reception of a
single person only—is described by Dr. Tucker, of Sligo, as follows:—
"It is built of stone and mortar, and brought to a round top. It is
sufficiently large for one person to sit on a chair inside, the door
being merely large enough to admit a person on his hands and
knees. When any of the old people of the neighbourhood, men or
women, are seized with pains, they at once have recourse to the
sweat-house, which is brought to the proper temperature by placing
therein a large turf fire, after the manner of an oven, which is left
until it is burned quite down, the door being a flat stone and air-
tight, and the roof, or outside of the house, being covered with clay,
to the depth of about a foot, to prevent the least escape of heat.
When the remains of the fire are taken out, the floor is strewn with
green rushes, and the person to be cured is escorted to the bath by
a second person carrying a pair of blankets. The invalid, having crept
in, plants himself or herself in a chair, and there remains until the
perspiration rolls off in large drops. When sufficiently operated on,
he or she, as the case may be, is anxious to get out, and the person
in waiting swaddles him up in the blankets, and off home, and then
to bed. I have heard old people say that they would not have been
alive, twenty years ago, only for the sweating-house.... Remains of
the Tig Allui are also found in the county Tyrone, of the following
dimensions:—five feet in height, nine in length, and four in width,
being built of solid masonry, and shaped like a bee-hive at the top."
Another, and a very important step in the progress of the bath was
the contrivance of a mode of heating by means of which the
temperature might be made uniform, and might be regulated in any
manner that should be required. The hot stones of the North
American medicine-man were clearly a very bungling and uncertain
expedient; little better than the warm skin of a newly-killed animal;
and the wood fire of the Irish sweating-houses was more
objectionable still, not only on account of the impossibility of
regulating the heat, but also from the resulting impurity of the
atmosphere and the danger of leaving fragments of the heated
ashes on the floor. The next contrivance, and that which has
continued to be the practice up to the present day, was the
construction of a furnace under the floor—in other words, a
hypocaust. Mr. Urquhart, speaking of the existence of baths among
the Mexicans and their probable introduction by the Phœnicians,
remarks:—"However magnificent their public monuments," their
baths were "such as are found in almost every house in Morocco,—a
small apartment seven feet square, with a cupola roof five to six feet
high, and a slightly convex floor, under one side of which there is a
fire, and a small low door to creep in by."
Reviewing the probable rise and progress of the bath, there seems
little doubt that the bath took its origin in the East, the dwelling-
place of our first parents, the birth-place of civilization and
knowledge. It was known at a very early period in Phœnicia. Mr.
Urquhart, in his recent work, "The Lebanon,"[4] relates his discovery
of a Phœnician temple, or crypt, among the ruins of Baalbeck, or
Baalbeth, the House of Baal—the Heliopolis, or City of the Sun, of
the Greeks; in which were traces of the existence of the bath. "But
the Phœnician crypt was not my only discovery. In a gap opening a
few feet into the masonry, I found mortar hard as stone where
exposed to the air, but soft within. Yet it was unlike other mortar; it
was dark grey, with particles of charcoal; when I brought out some,
it was recognised at once, and called kissermil, or ashes from the
bath. Those ashes are still used in this country for mortar, which
with this addition becomes as hard as stone. According to the old
construction, the baths were heated as an oven is, brushwood and
dung being used as well as wood. The combustion not being
complete, there remain various chemical compounds, alkali,
ammonia, sulphate and carbonate of lime, and carbon, which by
entering into new combinations, bind the mortar into a distinct
substance."
"One thing is clear, there were baths at Baalbeck. In the elaborately
finished bath of Emir Beshir at Ibtedeen, one peculiarity struck me
as evidencing their high antiquity in this land. It was the absence of
cocks; instead of which simple plugs or clots of cloth were used for
the pipes which brought the water into the basins. As the Romans
and Greeks used cocks, the art of the bath had not been derived
from them, but traced beyond them. Still it was curious to observe
these ashes in the midst of Cyclopic blocks. And yet why should not
the bath have belonged to the very earliest period of human society?
It is sufficiently excellent to be from the beginning."
"I remembered that in opening up the pavement of an ancient bath
on the western coast of Africa, I had come upon a somewhat similar
deposit, in large quantities, under the floor. This was gazul, the
product of a certain mountain in Morocco, resembling soapstone, but
composed of an admixture of silex, alumina, magnesia, and lime,
and which has the peculiar property of polishing the skin when
rubbed upon it, and so cleaning off the dead epidermis. Being used
for this purpose largely in the baths, the grey deposit under the ruin
in question is easily accounted for. Might not this same gazul, mixed
with kissermil, have been the deposit which I took for mortar at
Baalbeck?"
For the purpose of removing the dead epidermis from the surface of
the skin, "four processes have been adopted throughout the families
of the human race, and in successive times. The simple, the natural,
the first hit upon, was the rubbing down with the ball of the hand,
which is still the process used in this country for currying horses of
high breed. The three others, of a more refined and, I may say,
historical character, are, scraping, rolling, and polishing. The scraping
is with the strigil, which we know of from the Romans and Greeks,
but which is figured on the tombs of Lycia, and the Roman name of
which is derived from Mauritania. The rolling is that which we see to-
day practised by the Turks. The polishing is with the gazul, and
practised by the Moors, to whom it is confined, and who alone
possess the admirable substance which is used for it. Now, if gazul
was used by the early inhabitants of Baalbeck, their bathing process
belonged to the last of these systems, and they carried on a traffic
with Morocco."
From Phœnicia, from the coast of Tyre and Sidon, a knowledge of
the bath may have spread along the southern coast of the
Mediterranean, through Egypt, Tripoli, and Algiers, to Morocco and
the Pillars of Hercules; or it may, as Mr. Urquhart suggests, have
been earliest in use among the nations of Mauritania, and have been
carried by the Moors into the countries of the East. From Phœnicia,
the knowledge of the bath may have followed the line of caravan
communication into Russia, Persia, China, and Hindostan; while the
ships of the then greatest maritime country in the world would have
carried it to Greece, to Ireland, and to America. The bath is a
common practice in Russia; it is also well known in Persia,
Hindostan, and China; and, as we have already seen, its use in
North America, in Mexico, and Ireland, probably dates back to a very
early age. Its progress in Europe we shall presently see.
Speaking of the mode of heating the bath in Mexico and Morocco, I
have used the word hypocaust; this word is of Greek origin, and
signifies under-fire—that is, the fire is placed under the thing to be
heated; for example, under the foundation of the bath or of the
house. The Greeks and the Romans had no other means of heating
their houses than this; there was no open fire, but a fire under the
foundation, from which flues were carried upwards in the walls of
the building. When a great heat was required, as in the baths, the
foundation was supported on short columns (pilæ), and the entire
space between the columns was occupied with fire, while numerous
ascending flues distributed the heat around the rooms. Now it is
curious to find that at the present hour the Chinese continue the
same means of heating their houses.
That they also employ the sudatory process of bathing, is shown by
the following extract from Mr. Henry Ellis's "Journal of an Embassy to
China," published in 1817:—
"Near this temple (at Nankin) is a public vapour bath, called, or
rather miscalled, the Bath of Fragrant Water, where dirty Chinese
may be stewed clean for ten chens, or three farthings; the bath is a
small room of one hundred feet area, divided into four
compartments, and paved with coarse marble; the heat is
considerable, and as the number admitted into the bath has no limit
but the capacity of the area, the stench is excessive; altogether, I
thought it the most disgusting cleansing apparatus[5] I had ever
seen, and worthy of this nasty nation."
The Baths of Greece are celebrated for their magnificence; they
formed parts of buildings of vast extent and grandeur, termed
Gymnasia. The gymnasium was an institution of the Spartans of
Lacedæmonia or Laconia, and spread thence to other parts of
Greece, and notably to the metropolis of Attica, the famed city,
Athens. The gymnasium was sufficiently large to accommodate
several thousands of persons, and afforded space for the assembly
of philosophers, men of science, and poets, who delivered lectures
to their scholars and recited their verses; and for the pursuit of the
favourite games and exercises of their youths and men—namely,
leaping, running, throwing the disc or quoit, and wrestling; the
purpose of these exercises being to give strength to the people and
make them accomplished warriors.
The different parts of a gymnasium or palæstra, were as follows:—
1. The Porticos, in which were numerous rooms furnished with seats
for the professors and their scholars.
2. The Ephebeum, a large space in which the ephebi or youths planned
and practised their exercises.
3. The Apodyterium, or undressing room; also called Gymnasterium,
or the room for becoming nude.
4. The Elaiothesium, or anointing room, which was equally used by
those who were preparing for exercise, and those who had
completed their bath.
5. The Konisterium, or dusting room, where the bodies of the
wrestlers and other athletæ, after being anointed, were well dusted
over; probably as a defence to the skin against injury.
6. The Palæstra, or wrestling courts, which were bedded with sand
more or less deep, like the modern circus, in order to break the fall
of the combatants when they were thrown to the ground.
7. The Sphæristerium, or court for ball exercise and raquets.
8. The Peristyle, or Piazza, within which was the area of the Peristyle,
for walking, and the exercises of leaping, quoits, ball, and wrestling.
9. Then there were Xysti, or covered courts, for the use of the
wrestlers in bad weather; Xysta, which were walks between walls
open at the top and intended for hot weather; and a Xystic Sylvis, or
forest; the intervals of the numerous ornamental columns of the
building being so called, and being devoted to walking exercise.
10. Next came the Baths, which were hot, cold, and tepid water
baths; and a stove, or Laconicum, named after the city of Laconia
and the Lacedæmonians, from whom the Athenians derived their
knowledge of the hot-air bath.
11. And lastly, there was the Stadium, a segment of an ellipse, which
received its name from being one hundred paces long, equal to six
hundred feet, or something less than an eighth of a mile. The
Stadium was furnished with rows of seats for spectators, and was
intended for the exhibition of feats of running and exercises upon a
large scale.
The most remarkable Stadium known was one erected by Lycurgus
on the banks of the river Ilissus. It was built of Pentellick marble,
and was so magnificent a structure, that Pausanias the historian, in
describing it, informs his readers that they would not believe what
he was about to tell them, "it being a wonder to all that beheld it,
and of that stupendous bigness that one would judge it a mountain
of white marble."
There were several gymnasia in Athens, the most noteworthy being,
the Lyceum, the Academia, and the Cynosarges.
The Lyceum, founded on the banks of the river Ilissus, was
consecrated to Apollo; and not without reason, says Plutarch, but
upon a good and rational account, since from the same deity that
cures our diseases and restores our health, we may reasonably
expect strength and ability to contend in our exercises. The Lyceum
is also interesting to us as being the institution in which Aristotle
taught philosophy. Aristotle was wont to lecture to his scholars while
walking, and his disciples were therefore called Peripatetics; he
continued his teaching daily until the hour of anointing, which, with
the Greeks, was a preparation for dinner.
The Academia was situated in the suburbs of the city, on a piece of
ground that had been reclaimed from the marsh by draining and
planting. It was called after an old hero named Academicus. Plutarch
informs us that it was beset with shady woods and solitary walks fit
for study and meditation; in witness whereof another writer says:—
"In Academus' shady walks;"
and Horace writes:—
"In Hecademus' groves to search the truth."
Plato taught philosophy in the Academia; but having in consequence
of the unhealthy nature of the soil caught the ague, he was advised
to relinquish it for the Lyceum. "No!" said the old man, "I prefer the
Academy, for that it keeps the body under, lest by too much health it
should become rebellious, and more difficult to be governed by the
dictates of reason; as men prune vines when they spread too far,
and lop off the branches that grow too luxuriant."
The Cynosarges was also in the suburbs of Athens, not far distant
from the Lyceum. It was dedicated to the god of strength, Hercules;
and was interesting from its admission of strangers, and half-blood
Athenians. Its name is derived from the circumstance of a white dog
seizing upon a part of the victim that was being sacrificed to
Hercules by Diomus; and was the origin of the sect of philosophers
known as the "Cynics."
Baths of Rome.—When Greece was subjugated by the Romans, the
Romans carried back with them to Italy the taste for the bath. They
erected thermæ of great magnificence, and in so great number, that
at one period there were nearly nine hundred public baths in Rome.
Agrippa alone is said to have built one hundred and sixty, while
Mecænas has the credit of possessing the first private bath. The
most famed of the public baths were those of Titus, Paulus Æmilius,
Diocletian, Caracalla, and Agrippa. In these baths was centred all
that was most perfect in material, elaborate in workmanship, elegant
in design, and beautiful in art. Nothing was thought too grand or too
magnificent for their decoration. Superb marbles brought from the
most distant parts of the world; the choicest selections from the
riches of their conquests, the curious and wonderful in nature and in
art; precious gems and metals; and the finest works of the painter
and the sculptor. That beautiful production of the sculptor's art, the
Laocoon, was discovered among the ruins of the Baths of Titus, and
the celebrated Farnese Hercules in those of Caracalla.
The Baths of Agrippa were constructed of brick coated with enamel.
Those of Nero were supplied with water from the sea, as well as
fresh water. The Baths of Caracalla were a mile in circumference;
they possessed two hundred marble columns, sixteen hundred seats
of marble, and were capable of accommodating nearly two thousand
persons; while those of Diocletian surpassed all others in grandeur,
and occupied 140,000 men for many years in their construction.
Within the bath was collected all that contributed to the enjoyment,
the luxury, and the gaiety of existence of the Romans. Here they
practised their games, their athletic sports; here they came to learn
the news of the day, to listen to recitations of poetry and prose,[6] to
hear the eloquent harangues of their orators, and to be entranced
with the chords of melodious music. There were temples devoted to
dancing, to refreshment, to the bath; and in their abundant
gratitude they raised up appropriate statues to the gods who were
supposed to preside over their several enjoyments. The great hall of
their bath was ornamented with the statues of Hercules, the god of
strength; Hygeia, the goddess of health; and Æsculapius, the god of
medicine.
It is not to be wondered at, that, reared in the midst of the luxury, in
the enjoyment of their Balneæ or Thermæ, the Romans should have
carried with them their longing for the bath wheresoever they went,
wheresoever their victorious armies forced themselves a way; and
that, possessing a mastery over England and Wales, which they
maintained for nearly four hundred years, they should have founded
baths in their chief settlements in this country. Thus we have
remains of Roman baths in London, in Chester, in Bath, at Wroxeter
(Uriconium) in the neighbourhood of Shrewsbury, at Cirencester
(Corinium), at Carisbrooke, in Colchester (Camulodunum), and in
several other places besides.
But here we are compelled to draw a line of distinction between
those grand institutions of their own metropolis, which comprised, as
I have just described, their places of recreation, of exercise, of
amusement, of diversion, as well as their temples of health; which
were, in fact, a centralization of almost all the public institutions of
their city into one—and those particular parts of these institutions
which were specially devoted to health. Although the remains of the
Roman thermæ in England are large, although their construction
evinces a great perfection in many of the arts of social life,
particularly in the manufacture of bricks and pottery, yet they
scarcely bear comparison with the grander thermæ of ancient Rome,
and for this reason: that all that was simply ornamental—or, if I
might be permitted to say so, that was superfluous—has been
omitted, and nothing but the substantial and the wholesome allowed
to remain—that portion, in fact, which was purely devoted to health
and strength. We thus prune the bath down to its simpler elements,
and we prepare the way for the consideration of the bath as it has
been revived amongst us at the present day.
Without some explanation, it would be difficult to understand how
an institution which was regarded with so much veneration by
ancient Rome, should have totally fallen into decay in modern Rome;
and that the thermæ shall have ceased to have an existence in
Rome at the present day. It is clear that the games which were once
played, and the exercises which were practised, within the narrow
limits of the thermæ, grand though they were, have now sought a
wider sphere: the paintings of their great artists have been gathered
into the ecclesiastical edifices and academies; the statues and
sculpture have found their way into museums, or have been applied
to the decoration of modern palaces; refreshment is more
conveniently obtained in the cafés and restaurants; music and
singing have been transferred to the opera; recitation to the theatre;
poetry and prose to the library; and dancing to the assemblies; nay,
the great hall of the Bath of Agrippa is now, in all its integrity, a
place of Christian worship. In a word, the thermæ has become
decentralized; whether as the result of the adoption of foreign
fashions, or as a matter of convenience, it may be difficult to say;
but so it is, and nothing of it now remains but the bath—that temple
of the ancient thermæ over which Hercules, and Hygeia, and
Æsculapius presided of old, and over which (in its humbler shape)
they will continue to preside to the end of time.
The Baths of Titus have fortunately preserved to us a drawing, taken
from its walls, which illustrates the construction and the mode of
taking the bath among the Romans.
Beneath the bath is shown the furnace, or hypocaustum, for heating
the rooms, as also the water used in the latter stages of the process.
Then follows a series of rooms, of which the principal are:—
1. The Apodyterium, Gymnasterium, or Vestiarium: the undressing and
dressing-room.
2. The Tepidarium, which is warmed to a moderate temperature, and
is intended to prepare and season the body before entering the
hotter apartment.
3. The Caldarium, or Calidarium, sometimes called the Sudatorium, and
in the figure (Fig. 2), concamerata sudatio, was a room of higher
temperature, in which the perspiratory process was accomplished. In
this apartment there was commonly a recess, of a higher
temperature still, which was intended for special purposes, and was
named Laconicum, in compliment to the Spartans of Laconia.
4. After the Calidarium followed a Lavatorium (Lavatrina, Latrina),
called in the figure Balneum, in which the body was washed after
the process of perspiration was complete. The mode of washing was
to sit on the everted edge or lip of a large marble trough—the
labrum—and to be rinsed with warm water poured over the body by
means of a cup or small basin (pelvis).
5. The bather then went into the Frigidarium, where he received an
affusion of cold water, and where he reclined, or sat, or walked
about, until he was cool or dry.
6. From the Frigidarium the bather passed into the Elaiothesium, or
anointing room, where he was smeared with fragrant oils previously
to resuming his dress in the Vestiarium.
Besides these, which were the principal rooms, there were others
devoted to additional processes, such as shaving, hair-cutting,
depilation, and hair-plucking.
The Romans carried the indulgence and decoration of their baths to
so unreasonable a pitch of luxury and extravagance as to call forth
State restrictions upon their use, and the reproof of their
philosophers. Juvenal levels a shaft of satire against those who make
the bath the instrument of gluttony; and Pliny scolds the doctors for
declaring that the bath assists digestion, and for withholding their
denunciations against its excessive abuse. Moreover, the Emperor
Titus is said to have lost his life through excess of the bath, having
spent in it many hours of the day. We cannot, therefore, be
surprised that the time devoted to bathing should be limited by
imperial edict, as happened in the reign of the Emperor Hadrian, the
hours when the bath was open to the public being confined to two,
—namely, from three until five.
Pliny the Consul, in his admirable letters, speaks in most affectionate
language of the bath. "How stands Comum" (meaning Como, his
birthplace), says he, "that favourite scene of yours and mine? What
have you to tell me of the firm yet soft gestatio,[7] the sunny bath?"
In another letter, addressed to a lady, he says:—"The elegant
accommodations which are to be found at Narnia ... particularly the
pretty bath."
Describing his winter villa, Laurentium, after painting a series of
rooms, he continues:—
"From thence you enter into the grand and spacious cooling-room
belonging to the baths, from the opposite walls of which two round
basins project, large enough to swim in. Contiguous to this is the
perfuming-room, then the sweating-room, and beyond that the
furnace which conveys the heat to the baths. Adjoining are two
other little bathing-rooms, which are fitted up in an elegant rather
than costly manner: annexed to this is a warm bath of extraordinary
workmanship, wherein one may swim, and have a prospect at the
same time of the sea. Not far from hence stands the tennis-court,
which lies open to the warmth of the afternoon sun."
"Between the garden and this gestatio runs a shady walk of vines,
which is so soft that you may walk barefoot upon it without any
injury."
Alluding to the mode of life of one of his friends, he observes:—
"When the baths are ready, which in winter is about three o'clock,
and in summer about two, he undresses himself, and if there
happens to be no wind, he walks for some time in the sun. After this
he plays a considerable time at tennis, for by this sort of exercise,
too, he combats the effects of old age. When he has bathed, he
throws himself upon his couch till supper[8] time."
Seneca reproves the extravagance and self-indulgence of his
countrymen in a memorable letter (his eighty-sixth), which is as
follows:—
"I write from the very villa of Scipio Africanus, having first invoked
his manes, and that altar which I take to be the sepulchre of so
great a man.
"I behold a villa built of squared stone; the wall encloses a wood,
and has towers after the style of a fortification; the reservoir lies
below the buildings and the walks, large enough for the use of an
army; the bath is close and confined, dark, after the old fashion, for
our forefathers united heat with obscurity.
"I was struck with an inward pleasure when I compared these times
of Scipio with our own. In this nook did that dread of Carthage—to
whom our city owes her having been but once taken—wash his
limbs, wearied with labour; for, according to the ancient custom, he
tilled his fields himself. Under this mean roof did he live—him did this
rude pavement sustain.
"But who at this time would submit to bathe thus? A person is held
to be poor and sordid, whose house does not shine with a profusion
of the most precious materials, the marbles of Egypt being inlaid
with those of Numidia; unless the walls are ornamented with an
elaborate and variegated stucco, after the fashion of painting; unless
the chambers are covered with glass; unless the Thasian stone,
formerly a curiosity worthy of being placed in our temples, surrounds
the pools into which we cast our bodies weakened with immoderate
sweating; unless the water is conveyed through silver pipes.
"As yet, I have confined my remarks to private baths only. What
shall I say when I come to our public baths? What a profusion of
statues. What a number of columns do I see supporting nothing; but
placed as an ornament, merely on account of their expense. What
quantities of water murmuring down steps. We are come to that
pitch of luxury, that we disdain to tread upon anything but precious
stones.
"In this Bath of Scipio are small holes rather than windows, cut
through the wall, so as to admit the light without interfering with its
resemblance to a fortification.
"But now we reckon a bath fit only for moths and vermin, whose
windows are not so disposed as to receive the rays of the sun during
his whole career; unless we are washed and sunburnt at the same
time; unless from the bathing-vessel we have a prospect of the sea
and land. In fact, that which excited the admiration of mankind,
when first built, is now rejected as old and useless. Thus it is that
luxury finds out something new in which to obliterate her own
works.
"Formerly, baths were few in number, and not much ornamented; for
why should a thing of common life be ornamented, which was
invented for use, and not for the purposes of elegance? The water in
those days was not poured down in drops like a shower, neither did
it run always as if fresh from a hot spring; nor was its clearness
considered a matter of consequence. But, ye gods! what pleasure
was there in entering those obscure and vulgar baths when prepared
under the direction of the Cornelii, of Cato, or of Fabius Maximus?
For the most renowned of the ædiles had, by virtue of their office,
the inspection of those places where the people assembled, to see
that they were kept clean and of a proper and wholesome degree of
temperature; not of a heat like that of a furnace, such as has been
lately found out, proper only for the punishment of slaves convicted
of the highest misdemeanors. We now seem to make no distinction
between being warm and burning.
"How many do I hear ridiculing the simplicity of Scipio, who did not
admit the day into his sweating-places, or suffer himself to be baked
in a hot sunshine. Unhappy man! He knew not how to enjoy life!
"The water he washed in was not clear and transparent, but, after
rain, even thick and muddy. This, however, concerned him but little;
he came to the bath to refresh himself after his labour, not to wash
away the perfumes of a pomatumed body. What think you some will
say of this? I envy not Scipio: he lived in exile indeed, who bathed in
this manner.
"Should you be told further, that he bathed not every day—for those
who relate to us the traditions of early times, say that our
forefathers bathed their whole bodies on market-days only—it will be
answered, Then they were very uncleanly. How, think ye, they
smelt? Like men of labour and fatigue.
"Since dainty baths have been invented, we are become more nasty.
Horace, when describing a man infamous for his dissipation, what
does he reproach him with? With smelling of perfumed balls!
'Pastillos Rufillus olet.'"
Of the ancient Roman bath in England, we have several examples,
the most interesting being that which has been lately brought to
light in the ancient Roman city, Uriconium, in the neighbourhood of
Shrewsbury. Uriconium is close to the village of Uroxeter, commonly
pronounced Wroxeter, five miles from Shrewsbury. It is situated on
the property of the Duke of Cleveland, and is known to have existed
at the beginning of the second century of the Christian era, when
the Romans held dominion over England, and when England was a
part, and a highly-treasured part, of the Roman Empire.
It was of considerable size, having a boundary wall three miles in
circumference, and was, doubtless, a flourishing city; but fell a
victim to the ravages of fire and the sword during the fifth century,
and has since lain buried and unnoticed until the last few years,
when a society was formed for the purpose of excavating it.
The walls of the houses are remarkable for their thickness, namely,
three feet; while that of the wall of the town is four feet. They are
constructed upon a plan commonly adopted by the Romans—
namely, a facing of stone on each side, and the space between filled
in with rubble and that remarkable stone-like enduring mortar which
has suggested the name of a better kind of cement of the present
time, known as "Roman cement." The height of the houses was
thirty feet; but they had no upper storey, and there is no trace of
staircase.
Of the mode of warming the houses and, par excellence, the baths,
the Rev. Thomas Wright[9] observes:—"The Romans did not warm
their apartments by fires lighted in them.... The floor of the house,
formed of a considerable thickness of cement, was laid upon a
number of short pillars formed usually of square Roman tiles placed
one upon another, and from two to three feet high. Those of the
largest hypocausts yet found at Wroxeter were rather more than
three feet high. Sometimes these supports were of stone, and in one
or two cases in discoveries made in this country they were round.
They were placed near to each other and in rows, and upon them
were laid, first, larger tiles, and over these a thick mass of cement,
which formed the floor, and upon which the tesselated pavements
were set. Sometimes small parallel walls forming flues instead of
rows of columns supported the floor.... Flue tiles—that is, square
tubes made of baked clay with a hole on one side or sometimes on
two sides—were placed against the walls endways one upon another
so as to run up the walls."
My friend Mr. George Witt, having recently visited the ruins of the
ancient Roman bath still existing at No. 117, Bridge-street, Chester,
describes it as follows:—
"The most interesting of all the Roman antiquities of this ancient city
are the remains of a private Roman bath, showing the Hypocaustum,
or heating place beneath, in a state of great preservation. The
hypocaust is 18 feet long by 8 feet wide, and 3 feet high. The roof
was supported on thirty-two stone pillars (of a single block), broader
at the base and the top, and narrower in the middle; of these
twenty-eight still remain. On the top of the columns are placed, by
way of capitals, strong tiles from 17 to 23 inches square, and 3
inches thick, reaching from pillar to pillar, thus forming, at the same
time, the roof of the hypocaust and the floor of the room above.
Over all these is a bed of hard concrete, 9 or 10 inches thick, the
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about books and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebookball.com