I study how bacteria build and rebuild their cell wall, and how the enzymes that reshape pep­ti­do­gly­can help a cell cope with stress and antibiotics.

About me

I am a bacterial mol­ec­u­lar microbiologist. I did my PhD at the CSIR-Centre for Cellular and Mol­ec­u­lar Biology (CCMB) in Hyderabad, and I have spent the past three years as a postdoctoral Research Associate at the University of Sheffield, working with col­lab­or­at­ors at the Universities of Oxford and Birmingham.

My work is on how bacteria remodel their cell en­vel­ope. During my PhD I identified LdtF, the enzyme that releases Braun's lipo­pro­tein from pep­ti­do­gly­can in E. coli. Afterwards, as a Project Associate at CCMB, I studied how the wall is remodelled during cold shock. In Sheffield I moved to the l,d-trans­pep­ti­dases of Rhizobium, where we found l-amino acids built into pep­ti­do­gly­can and an enzyme whose reaction changes with pH.

I combine bacterial genetics, re­com­bi­nant protein biochemistry, enzyme assays and LC–MS/MS-based pep­ti­do­gly­com­ics to study enzyme function, cell-wall ar­chi­tec­ture and how bacteria adapt, with relevance to antibiotic susceptibility.

Enzymes I have worked on

LdtF cuts the tether Lys mDAP3 3 LPS OM PG Lpp

LdtF

Escherichia coli

A nonconformist enzyme in the LDT family: it cuts Braun's lipo­pro­tein from pep­ti­do­gly­can.

Usual growth 4→3 cross-links Cold shock 3→3 cross-links (LdtD)

LdtD

Escherichia coli

Shows how E. coli changes its own jacket during cold shock.

acyl-enzyme Cys–S~mDAP3 pH sets the preference transpeptidation cross-link to a neighbouring stem carboxypeptidation tripeptide, with water as acceptor substitution D- or L-residue at position 4 endopeptidation an existing cross-link is cut

Rj LDTs, LdtRj8 and LdtCd3

Rhizobium johnstonii, Clos­trid­i­oi­des dif­fi­cile

The Rj LDTs are multitasking enzymes that carry out several pep­ti­do­gly­can-re­mod­el­ling reactions, breaking the stereotype of one enzyme, one reaction.

LdtRj8 can change the stereo­chem­is­try of pep­ti­do­gly­can.

LdtCd3 supports our finding that l-amino acid in­cor­por­a­tion at the C-terminus of the stem peptide is a broader phenomenon.

Research vision

The bacterial cell wall is usually drawn as a fixed scaffold, but its chem­is­try changes all the time. Enzymes cut, add and exchange residues, and this editing is how a cell keeps its en­vel­ope intact in the cold, in acid, under antibiotics and inside a host. I want to know what role cell-wall enzymes play under stress con­di­tions, well enough to predict the wall a cell will build or modify. This matters to me for two reasons. One is to work towards combination therapy against antimicrobial resistance (AMR). The other is that these enzymes are what lets a symbiont remodel its en­vel­ope as it settles into its host.

Research domains

My work has two strands: the enzymes that remodel the wall, and the chem­is­try of the wall itself.

Domain 1, PhD

Enzymes that remodel the cell wall and hold the en­vel­ope together

A two-state cycle. On the left, Braun's lipoprotein links the outer membrane to a stem peptide in the sacculus. On the right, the link has been cut: the lipoprotein stays in the membrane with its C-terminus free and the stem is a tripeptide. LdtF drives the cutting reaction, LdtA, LdtB and LdtC drive the attaching reaction. Below, the consequences of where the balance sits, and a second LdtF reaction removing glycine from position 4.
Working model: the tether is re­ver­si­ble. LdtA, LdtB and LdtC attach Lpp to mDAP3; LdtF cuts the same bond. The balance between the two reactions sets how much of the Lpp pool is bound, how far the outer membrane sits from the sac­cu­lus, and how well the en­vel­ope holds up under stress. The same residue is also where glycine enters the stem and leaves it again. A working model drawn for this page, not a figure from the paper.

Braun's lipo­pro­tein is cut from the cell wall by an l,d-trans­pep­ti­dase paralog

Bahadur R, Chodisetti PK, Reddy M. PNAS, 2021. doi:10.1073/pnas.2101989118

Diagram of the Gram-negative cell envelope in three layers: outer membrane with lipopolysaccharide in its outer leaflet, peptidoglycan and inner membrane. Two glycan strands carry stem peptides joined by a 4-3 and a 3-3 cross-link. Braun's lipoprotein is shown bound, its C-terminal lysine attached to meso-diaminopimelic acid in a stem peptide, and free with its C-terminus exposed at the cell surface. Blue arrows mark the two activities of LdtF: cleavage of the peptidoglycan–lipoprotein cross-link, and removal of a terminal glycine.
Where LdtF acts in the en­vel­ope. (A) Lpp staples the outer membrane to the sac­cu­lus: the ε-amino group of its C-terminal lysine is amide-linked to mDAP3 of a stem peptide, while about two-thirds of Lpp stays free with its C-terminus exposed at the surface. Blue arrows mark the two activities of LdtF. (B) The disaccharide repeat, GlcNAc β-(1→4) MurNAc, with the stem peptide carried on the d-lactyl ether at C-3 of MurNAc. (C) Both reactions release a group from mDAP3 and return the stem to a tripeptide: the l,d-en­do­pep­ti­dase reaction un­teth­ers the outer membrane, and the glycine-specific car­boxy­pep­ti­dase reaction removes a Gly in­cor­por­at­ed at position 4. Drawn for this page; the biology follows Bahadur, Chodisetti & Reddy, PNAS 2021.

In E. coli, Braun's lipo­pro­tein (Lpp) is cova­lent­ly tethered to pep­ti­do­gly­can and ties the outer membrane to the cell wall. How that link is broken was unknown. I identified LdtF, a paralog of the l,d-trans­pep­ti­dases, as the enzyme that cleaves the Lpp–pep­ti­do­gly­can bond, and showed that it also has car­boxy­pep­ti­dase activity on glycine-containing pep­ti­do­gly­can.

This gave a function to an orphan member of the LDT family. It also gave the field an enzyme that releases outer-membrane protein from the cell wall, the reverse of the an­chor­ing reaction.

LdtD remodels pep­ti­do­gly­can in the cold

Nallamotu KC, Bahadur R, Kaul M, Reddy M. Journal of Bacteriology, 2023. doi:10.1128/jb.00382-22

Cells grown at low tem­per­at­ure change the chem­is­try of their cell wall. We showed that this cold-shock re­mod­el­ling depends on the l,d-trans­pep­ti­dase LdtD, which builds 3-3 cross-links in place of the usual 4-3 ones.

This links one enzyme to one en­vi­ron­men­tal stress. It also supports the view that the 3-3 cross-link is a con­di­tion­al, adaptive feature of the wall, not a constant one.

Domain 2, postdoctoral

Stereo­chem­is­try and pH as controls on cell-wall ar­chi­tec­ture and fitness

Three panels. A: a pea plant in soil with pink nodules on its roots and free-living rhizobia in the soil. B: an infected nodule cell containing branched bacteroids, each inside a symbiosome. C: the envelope of a bacteroid, with outer membrane, peptidoglycan and inner membrane; the peptidoglycan is marked as the layer remodelled by L,D-transpeptidases.
The system these enzymes work in. (A) Rhizobium johnstonii lives free in the soil and forms nodules on the roots of pea (Pisum sativum). (B) Inside an infected nodule cell it becomes a branched bacteroid, each one enclosed in a plant-made sym­bio­some. (C) The bacteroid en­vel­ope, with the pep­ti­do­gly­can layer that l,d-trans­pep­ti­dases remodel. These surroundings can be acidic, the range in which LdtRj8 is most active. Whether this re­mod­el­ling matters in the host is what I want to test. A schematic drawn for this page, not data.
A donor stem peptide loses its terminal D-alanine to form one acyl-enzyme intermediate on the catalytic cysteine. Four arrows lead from that intermediate to four reactions, chosen by the acceptor available: a neighbouring stem gives a cross-link, water gives a tripeptide, a free amino acid gives substitution, and an existing cross-link is cut. A pH axis shows the preference switching between pH 5 and 6. At the foot, a D-residue at position 4 leaves the stem a competent donor while an L-residue blocks further cross-linking.
Working model: one in­ter­me­di­ate, four fates. A single l,d-trans­pep­ti­dase makes one acyl-enzyme in­ter­me­di­ate, and whichever ac­cept­or reaches it first decides the reaction. pH changes which outcome wins, not just how fast the enzyme works. The stereo­chem­is­try of the residue installed at position 4 then decides whether the stem can still be used: with an l-residue there, it is a poor donor and further cross-linking slows. A working model drawn for this page, not a figure from the manuscript.

The wall carries both stereo­chem­is­tries at the end of the stem peptide

Rady BJ*, Bahadur R*, Evans CA, Mesnage S. bioRxiv, 2026. doi:10.64898/2026.06.30.735398. *Equal contribution. Revised manuscript in preparation.

Pep­ti­do­gly­can is built with d-amino acids at the end of the stem peptide. That stereo­chem­is­try is what sets it apart from protein, and what many of its enzymes are assumed to require. Working on Rhizobium johnstonii, we found pairs of muro­pep­tides with the same mass but different re­tent­ion times. Mass spec­trom­e­try could not tell them apart. NMR could: the two differ only in the con­fig­u­ra­tion of the terminal alanine, so a sub­stan­tial fraction of this wall ends in l-Ala rather than d-Ala.

Screening all 18 l,d-trans­pep­ti­dases of R. johnstonii in a purpose-built E. coli Lemo21(DE3) Δ6ldt host identified the enzyme responsible. LdtRj8 cat­a­lys­es every reaction described for the family, and which one it favours shifts with pH. In sub­sti­tu­tion it accepts all 20 d-amino acids and most l-amino acids, and the same is true of LdtCd3 from Clos­trid­i­oi­des dif­fi­cile, so this is a property of the family, not of a single enzyme.

What it means for fitness

The sub­sti­tu­tion changes how the wall is built. Replacing the terminal d-Ala with l-Ala cuts sub­se­quent 3-3 cross-linking from 51% to 12% for LdtRj8, and from 33% to 1.2% for LdtCd3. An l-residue at that position is therefore a local brake on cross-linking. It gives the cell a way to control how tightly a patch of wall is cross-linked, using its large pool of l-amino acids instead of scarce d-amino acids. LdtRj8 works best at about pH 5–6. A rhizobium meets that range in the rhizo­sphere, in root hairs and inside the sym­bio­some, where it has to remodel its en­vel­ope while it colonises the host.

Where I want to take it

I want to take this in two directions. The first is to ask whether l-amino-acid sub­sti­tu­tion changes fitness in the host, by testing whether a strain that cannot make it is worse at col­on­isa­tion, at nod­u­la­tion, or at surviving acid stress. The second is anti­bac­terial. In several pathogens, the 3-3 cross-links made by l,d-trans­pep­ti­dases are how the cell escapes β-lactams, so an enzyme that can be pushed into sub­sti­tu­tion instead of cross-linking, either by pH or by the amino acid available to it, is worth testing on that route. The E. coli Δ6ldt strain I built for this work secretes the enzyme into the medium, so it is also a practical way to produce peri­plas­mic proteins that are otherwise hard to express.

Other con­tri­bu­tions

Methods de­vel­op­ment

Developed an ultra­cen­tri­fuge-free PG isolation workflow in 96-well plates, PG detection by western blotting, in-house mu­tano­ly­sin production, and paired RP-HPLC and LC–MS methods.

Pep­ti­do­gly­can hy­dro­lase biology

Char­ac­ter­ised the murein en­do­pep­ti­dase MepR, which hy­dro­lys­es 3-4 cross-links in an­hy­dro­muro­pep­tides (unpublished work).

Selected publications

  1. Rhizobial enzyme reveals pH-driven catalytic switching and l-amino acid in­cor­por­a­tion by l,d-trans­pep­ti­dases

    Rady BJ*, Bahadur R*, Evans CA, Mesnage S. bioRxiv, 2026. doi:10.64898/2026.06.30.735398

    One enzyme, LdtRj8, cat­a­lys­es every reaction described for the family and changes its preferred reaction with pH.

  2. Pep­ti­do­gly­can remodeling by an l,d-trans­pep­ti­dase, LdtD during cold shock in E. coli

    Nallamotu KC, Bahadur R, Kaul M, Reddy M. Journal of Bacteriology, 2023; 205(1): e0038222. doi:10.1128/jb.00382-22

    Cold-shock re­mod­el­ling of the wall depends on LdtD, which builds 3-3 cross-links.

  3. A LytM-domain factor, ActS, functions in two distinctive pep­ti­do­gly­can hydrolytic pathways in E. coli

    Chodisetti PK, Bahadur R, Amrutha RN, Reddy M. Frontiers in Mi­cro­bi­ol­o­gy, 2022; 13: 913949. doi:10.3389/fmicb.2022.913949

  4. Cleav­age of Braun's lipo­pro­tein Lpp from the bacterial pep­ti­do­gly­can by a paralog of l,d-trans­pep­ti­dases, LdtF

    Bahadur R, Chodisetti PK, Reddy M. Proceedings of the National Academy of Sciences USA, 2021; 118(19): e2101989118. doi:10.1073/pnas.2101989118

    Identified LdtF as the enzyme that cuts Braun's lipo­pro­tein from pep­ti­do­gly­can.

* Co-first authors, contributed equally.

Research experience

2023–2026

Research Associate

University of Sheffield, UK

  • In­ves­ti­gat­ed PG re­mod­el­ling and bacterial en­vel­ope integrity, with emphasis on the l,d-trans­pep­ti­dases of Rhizobium johnstonii (formerly R. le­gu­mi­no­sa­rum) 3841.
  • Char­ac­ter­ised 18 putative l,d-trans­pep­ti­dases and in­ves­ti­gat­ed atypical amino-acid in­cor­por­a­tion into PG using PG isolation, RP-HPLC, LC–MS/MS, PG-Finder and mol­ec­u­lar and bio­chem­ic­al ap­proach­es.
  • Worked on joint projects with groups at the University of Oxford and the University of Birmingham.
2022–2023

Project Associate

CSIR-Centre for Cellular and Mol­ec­u­lar Biology (CCMB), India

  • In­ves­ti­gat­ed LdtD-mediated PG re­mod­el­ling during cold shock in E. coli.
2016–2022

PhD Researcher

CSIR-Centre for Cellular and Mol­ec­u­lar Biology (CCMB), India

  • Studied PG hydrolases and cell-en­vel­ope re­mod­el­ling in E. coli.
  • Identified and char­ac­ter­ised LdtF and its role in PG–Lpp cross-link cleav­age.
2014–2015

Scientific Officer

Bhabha Atomic Research Centre (BARC), Mumbai, India

Education

2022

PhD, Mol­ec­u­lar Mi­cro­bi­ol­o­gy

Jawaharlal Nehru University, New Delhi (CSIR-Centre for Cellular and Mol­ec­u­lar Biology), India

2012

MSc, Bio­tech­nol­o­gy

Chhatrapati Shahu Ji Maharaj University, Kanpur, India

2009

BSc

Chhatrapati Shahu Ji Maharaj University, Kanpur, India

Technical expertise

  • Bacterial genetics
  • Mol­ec­u­lar biology and re­com­bi­nant DNA
  • Protein expression and pu­ri­fi­ca­tion (bacterial and insect cell systems)
  • Pep­ti­do­gly­can isolation
  • Pep­ti­do­gly­com­ics
  • LC–MS/MS
  • RP-HPLC
  • PG-Finder and computational analysis
  • Enzyme kinetics
  • Mi­cros­co­py
  • Antibiotic and stress phe­no­typ­ing
  • Ex­per­i­men­tal design and collaboration

Awards, grants and recognition

  • 2025/2026Recognised as outstanding postdoc
  • 2025General Travel Grant, Bio­chem­ic­al Society
  • 2025External Event Grant, Mi­cro­bi­ol­o­gy Society
  • 2023–2026UK Global Talent Visa, Bioscience (UKRI-endorsed route)
  • 2022Best Poster Award, All India Cell Biology Conference

Conferences

  • 20258th Great Wall Symposium, Catania, Sicily, Italy (poster)
  • 20237th Great Wall Symposium, Vila Galé Sintra, Portugal (poster)
  • 202244th All India Cell Biology Conference, University of Kashmir, India (poster)
  • 2018International Congress of Cell Biology, Hyderabad, India (participant)

Contact

I am glad to hear from possible col­lab­or­at­ors, and from anyone working on pep­ti­do­gly­can.

Your message goes to Raj by email. Nothing is stored on this site.

Email [email protected]
ORCID 0000-0002-4673-6844
Sheffield, United Kingdom

References available on request.

Figure