فهم ترابط البكتين في جدران خلايا النبات Understanding pectin cross-linking in plant cell walls

المجلة: Communications Biology، المجلد: 8، العدد: 1
DOI: https://doi.org/10.1038/s42003-025-07495-0
PMID: https://pubmed.ncbi.nlm.nih.gov/39825091
تاريخ النشر: 2025-01-17

فهم ترابط البكتين في جدران خلايا النبات

إيرابونوسي أوبوميغي (1) لاين ج. برينتس بيتر لوين-جونز () فابيين باختيجر ناثان رامزي شيكو كيشي-إيتاكورا مارتن و. غولدبرغ تيم ج. هوكينز جيمس إي. سبراتلز هيذر نايت وغابرييل سي. سوسو
البكتين هو مكون رئيسي من جدران خلايا النباتات. مدى ترابط سلاسل البكتين مع بعضها البعض يحدد خصائص حاسمة بما في ذلك قوة جدار الخلية، المسامية، وقدرة الجزيئات الصغيرة ذات الأهمية البيولوجية على الوصول إلى الخلية. على الرغم من أهميته، لا تزال هناك فجوات كبيرة في فهمنا، على المستوى الجزيئي، لكيفية تأثير روابط البكتين المتقاطعة على الخصائص الميكانيكية والفيزيائية لجدران الخلايا. تستخدم هذه الدراسة نهجًا متعدد التخصصات، يجمع بين محاكاة الديناميكا الجزيئية، والتحقيقات التجريبية، والنمذجة الرياضية، لتوضيح آلية ترابط البكتين المتقاطع وتأثيره على مسامية جدار الخلية. تتحدى الجوانب الحاسوبية من هذا العمل نموذج صندوق البيض السائد، مفضلة بدلاً من ذلك نموذج السحاب لترابط البكتين المتقاطع، بينما يبرز عملنا التجريبي التأثير الكبير لترابط البكتين المتقاطع على مسامية جدار الخلية. يعزز هذا العمل فهمنا الأساسي للكيمياء الحيوية التي تدعم هيكل ووظيفة جدار الخلية النباتية. هذه المعرفة لها تداعيات مهمة على التكنولوجيا الحيوية الزراعية، حيث تخبرنا عن الخصائص الكيميائية للبكتينات النباتية التي تناسب تحسين مقاومة المحاصيل وقابليتها لاستخراج الوقود الحيوي من خلال تعديل جدار الخلية.
على عكس خلايا الحيوانات، فإن خلايا النباتات محاطة بجدار خلوي، يوفر القوة والحماية، بينما يسمح بالنمو. بالإضافة إلى دورها الهيكلي، تشكل جدار الخلية حاجزًا مهمًا يحدد ما يمكن وما لا يمكنه الوصول إلى الخلية. إنه مسامي، مما يسمح بمرور الماء والغازات، لكن درجة المسامية قد تختلف، وهذا ما يحدد أي الجزيئات الصغيرة يتم قبولها في الخلية أو يتم عرقلتها من الدخول. يتكون الجدار الخلوي الأولي للنباتات من كربوهيدرات معقدة تشمل البكتين والسليلوز والهيميسليلوز وبروتينات مدمجة ذات وظائف هيكلية وإنزيمية. تتميز البكتينات بوجود هيكل غني بحمض الجالاكتورونيك (GalA) وتشكل روابط تساهمية مع بعضها البعض. ، مما يؤدي إلى تشكيل الهيكل المعقد ثلاثي الأبعاد لجدار الخلية الذي يتحكم في العديد من الخصائص، بما في ذلك القوة الميكانيكية وخصائص المسامية للجدار أظهرت التقديرات الأولى لتمسّك جدران الخلايا أن الجدار قدّم حاجزًا أكثر أهمية لمرور الجزيئات الصغيرة مما كان يُفترض سابقًا، مما يشير إلى أن التحكم في المسامية يمكن أن ينظم التواصل الخلوي. في الوقت الحاضر، فإن فهم وتوقع حجم مسام جدار الخلية النباتية أمر حاسم في التكنولوجيا الحيوية الزراعية. يمكن أن تحد أحجام مسام جدار الخلية الصغيرة من دخول وفعالية إنزيمات تكسير جدار الخلية، مثل السليلوزات، التي تستخدمها مسببات الأمراض الغازية. ، التي تسبب محصولًا مذهلاً
الخسائر في جميع أنحاء العالم وعلى العكس، لنفس السبب، فإن الجدران المسامية أكثر ملاءمة لإنتاج الوقود الحيوي من خلال عملية التحليل السكري. .
تتأثر جيلاتين البكتين بتكوين الروابط المتقاطعة بين سلاسل البكتين أو سلاسل السكر الجانبية لها. تمثل بكتينات الهوموغالكتورونان (HG) الغالبية العظمى من بكتين جدار الخلية، وتتصل بشكل أساسي ببعضها البعض عبر أيونات الكالسيوم. ) بين مجموعات الكربوكسيل التي تم إزالة الميثيل منها (DM)، كما هو موضح في الشكل 1. يُعتقد أن آلية الربط هذه تؤدي إلى تجميع سلاسل HG من خلال ما يُعرف بنموذج “صندوق البيض”. على الرغم من أن العمل الأخير قد وضعت صلاحية هذا النموذج موضع تساؤل. يتم تحديد درجة DM للبيكتين من خلال عمل عائلة كبيرة من الإنزيمات، وهي إنزيمات ميثيل استيراز البيكتين (PMEs). وبواسطة العديد من مثبطات PME (PMEIs) يمكن أن يرتبط HG الذي تم إزالة الميثيل إستر منه أيضًا بالمجالات البكتينية الأخرى والزيلوجلوكانات. تتصل سلاسل الرامنوغالكتورونان II (RGII) الأكثر تعقيدًا ببعضها البعض عبر روابط ثنائية الإستر البوراتية بين سلاسل السكر الجانبية المجاورة. . في كل حالة، يؤثر درجة الربط المتقاطع على المسامية، أي المساحات من النانومتر إلى الميكرومتر داخل الجدار .
في هذا العمل، استخدمنا نهجًا متعدد التخصصات لتسليط الضوء على آلية الربط المتقاطع للبكتين ولإقامة صلة بين بنية البكتين ومسامية جدار الخلية.
الشكل 1 | آليات الربط المتقاطع المختلفة لسلاسل الهيدروجيل. المجموعات الوظيفية المختلفة المعنية، وهي مجموعات الكربوكسيل المؤينة (-COOH)، ومجموعات الكربوكسيل غير المؤينة (- )، تم تمييز المجموعات الكربوكسيلية الميثيلية (حمض الجالكتورونيك الميثيل-إستر) والمجموعات الكربوكسيلية الأسيتيلية بالأرجواني والأخضر والبرتقالي والوردي، على التوالي. تم اقتراح آليتين متميزتين لربط سلاسل الهيدروكسي جيل عبر الجسور: نموذج “صندوق البيض” و نموذج “السحاب”.
نموذج صندوق البيض يُوجد عادةً في الأدبيات المتعلقة بربط البكتين، ويتضمن مجموعتين هيدروكسيل (-OH، أزرق فاتح) ومجموعتين – المجموعات. هنا، نقدم نموذج السحاب (الذي يتضمن اثنين – فقط) بدلاً من ذلك في ضوء النتائج الحسابية التي تم مناقشتها في قسم “النتائج”. يمكن أن يحدث الربط المتقاطع لـ HG أيضًا من خلال الروابط الهيدروجينية، بشكل بارز بين مجموعات الكربوكسيل الميثيلية – أو حتى مجموعات الكربوكسيل المؤينة في ظروف حمضية جدًا.
على وجه التحديد، قمنا بالتحقيق في التفاصيل على المستوى الجزيئي لربط HG عبر المحاكاة الجزيئية، وقمنا بتحديد طاقة آليات الربط المختلفة. كما قمنا أيضًا بالتحقيق بشكل منهجي في علاقة الهيكل-الوظيفة بين تفعيل سلاسل HG وحركية تجميعها. تشير نتائجنا إلى أن نموذج “صندوق البيض” العريق قد لا يكون كافيًا لوصف -التشابك الوسيط في البكتين. للتحقيق في تأثير تشابك البكتين على مسامية جدار الخلية، استخدمنا نسخة موسعة من التقنية المعتمدة على المجهر الفلوري التي ابتكرها ليو وآخرون. من خلال تصور التغيرات في الفلورية من غشاء البلازما المعلم بالفلور على مر الزمن، قمنا بمراقبة ديناميات جزيء التثبيط لاختراق جدران الخلايا لنباتات الأرابيدوبسيس ثاليانا من النوع البري وطفرة تعاني من ضعف في الربط المتقاطع للبكتين في جدران الخلايا. لفهم نتائجنا التجريبية، استخدمنا النمذجة الرياضية المستندة إلى ديناميات السوائل. تتماشى نتائجنا مع البيانات التجريبية، وتوفر تفسيرًا للاعتماد الزمني على تدهور الفلورية من حيث شكل العينة، وتضع الأسس لتطوير نموذج تنبؤي في المستقبل – لقياس المسامية وربما التعقيد. جدار الخلية.

النتائج

المحاكاة الجزيئية

نموذج صندوق البيض لربط HG غير مدعوم. حيث أن أكثر المجالات البكتينية وفرة في جدران خلايا النباتات هو ، ركزنا على فهم التفاعلات بين سلاسل البكتين عالية الميثوكسيلي. كان هدفنا هو توضيح كيفية ارتباط البكتين على المستوى الجزيئي – مما قد يساعدنا بدوره على فهم درجات المسامية المختلفة داخل جدار الخلية. للقيام بذلك، قمنا بدراسة الطاقة لكل نوع من تفاعلات HG-HG: الخيارات الرئيسية في هذا الصدد ملخصة في الشكل 1. أولاً، نعتبر مجموعات الكربوكسيل من GalA، التي نتوقع في معظم الحالات أن نجدها في شكلها غير المؤين (-COO-)، حيث أن قيمة pKa لها منخفضة نسبياً (حوالي 3.6). في هذا السيناريو، اثنان – يمكن أن تتقاطع المجموعات على سلسلتين مختلفتين عبر جسر.
تشير الكثير من الأدبيات الموجودة إلى أن هذه الجسور الكالسيومية تتضمن أيضًا مجموعتين هيدروكسيل (-OH). هذه الإمكانية، الموضحة في الشكل 1، تؤدي إلى ظهور ما يسمى بنموذج “صندوق البيض” لربط البكتين. . هذا النموذج مقبول على نطاق واسع بين علماء النبات، ومع ذلك، من الصعب تجريبياً تحديد الموقع الدقيق و-الأهم-التنسيق لـ الأيونات داخل مناطق الربط المتقاطع لسلاسل الهيدروجيل .
تم استخدام تشتت الأشعة السينية بزاوية صغيرة (SAXS) سابقًا للتحقيق في التفاصيل الهيكلية لـ التشابك في البكتين والألجينات . ومع ذلك، من المهم أن نلاحظ أن SAXS لا يمكن أن يوفر دليلاً مباشراً لنموذج صندوق البيض. ومن الجدير بالذكر أن حتى التجارب الحديثة
العمل على الأرابيدوبسيس يستشهد بالجهود السابقة في النمذجة، وبشكل خاص عمل براكيني وآخرين. ، كدليل على نموذج صندوق البيض. من المثير للاهتمام، أن براكيني وآخرين لا تدعو إلى نموذج صندوق البيض الأصلي ولكن تقترح تكوين “صندوق البيض المنزاح” لسلاسل الجالاكتورونات والجولورونات.
في الواقع، قد أثارت الأعمال الأخيرة تساؤلات حول صحة هذا النموذج لصالح نموذج “السحاب”. ، كما هو موضح في الشكل 1، والذي يتضمن الاثنين – مجموعات فقط. هذه الفجوة الصغيرة المضللة بين نموذج صندوق البيض ونموذج السحاب تترجم في الواقع إلى تكوين مختلف تمامًا لسلسلة HG ككل – كما سنناقش بمزيد من التفصيل لاحقًا.
في الحالات التي يمكن أن توجد فيها مجموعة الكربوكسيل من GalA في شكلها المؤين، من المعقول توقع بعض درجة من الروابط الهيدروجينية بين مجموعتين -COOH تنتميان إلى سلسلتين مختلفتين، دون الحاجة إلى تضمين جسر . في الواقع، بينما يكون هناك كمية كافية من الكالسيوم، فإن تشكيل الجسور هي الآلية الرئيسية لربط HG، في البكتين الميثيلي و/أو الأسيتيل المفرط، يحدث الربط بفضل الروابط الهيدروجينية، كما هو موضح في الشكل 1.
وبالتالي، يتم تحديد المسامية الناتجة عن شبكة البكتين داخل جدار الخلية من خلال تنافس آليات الربط المتقاطع المختلفة لـ HG. من أجل تقييم استقرار هذه الخيارات المختلفة للربط المتقاطع لسلاسل HG، قمنا بإجراء محاكاة ميتا ديناميكية. يتم مناقشة التفاصيل بشكل موسع في قسم “الطرق”: باختصار، الميتا ديناميكا هي تقنية حسابية لتحسين العينة تستفيد من محاكاة الديناميكا الجزيئية (MD) لإعادة بناء ملفات الطاقة الحرة بالنسبة لواحد أو أكثر من “المتغيرات الجماعية” – في هذه الحالة، درجات الحرية المحددة التي تصف عملية الربط المتقاطع. للبدء، قمنا بحساب عدد التنسيق الجاري من خلال محاكاة غير متحيزة (أي، دون تطبيق طريقة الميتا ديناميكا حتى لا نتلاعب بالتطور الزمني الطبيعي للنظام) لـ في الماء (أي متوسط عدد جزيئات الماء التي يمكن العثور عليها ضمن مسافة معينة من الـ أيون) من أجل التحقق من موثوقية إعدادنا الحاسوبي. تشير النتيجة، الموضحة في الشكل 2a، إلى وجود غلاف أولي للترطيب ممتد إلى حد ما (ضمن 2.5 و Å ) التي تحتوي في المتوسط جزيئات الماء. هذا يتماشى مع النتائج السابقة التي تم الحصول عليها من خلال مجال القوة CHARMM بالإضافة إلى محاكاة الديناميكا الجزيئية من أولى المبادئ. .
من خلال محاكاة الميتا ديناميكا، قمنا بدراسة مكاسب الطاقة الحرة المرتبطة بتكوين الجسر، الذي نبلغ عنه في الشكل 2ب كدالة لكل من عدد التنسيق لـ مع الماء وعدد التنسيق لـ مع أي من ذرات الأكسجين الأربعة التي تنتمي إلى مجموعتي الكربونيل المزيلتين للبروتون (انظر الشكل 2d). تشير نتائجنا إلى أن قادر على الربط الكامل مع اثنين المجموعات فقط من خلال آلية “السحاب” المذكورة أعلاه، دون الحاجة إلى تضمين مجموعات -OH عبر آلية “صندوق البيض”. الرابط الأكثر استقرارًا يتضمن جميع ذرات الأكسجين الأربعة التي تنتمي إلى الاثنين المزيلين للبروتون.
الشكل 2 | محاكاة الميتا ديناميكا لربط HG. أ متوسط عدد التنسيق المتحرك لـ مع جزيئات الماء في محاكاة ديناميكية جزيئية غير متحيزة. تبرز المنطقة المظللة عدم اليقين المرتبط بتقديرنا (خطأ الحد الأدنى – الحد الأقصى)؛ يوضح الشكل الداخلي تنسيقًا سداسيًا. سطح الطاقة الحرة بالنسبة لتكوين وحدة واحدة – الجسر، المبلغ عنه كدالة لعدد التنسيق لـ أيون مع جزيئات الماء بالإضافة إلى رقم التنسيق لـ أيون مع أي من ذرات الأكسجين الأربعة التي تنتمي إلى الاثنين – مجموعات ملف الطاقة الحرة بالنسبة للتفاعل بين وواحد مجموعة، تم الإبلاغ عنها كدالة للمسافة بين أيون وأي من ذرتي الأكسجين التابعتين لـ مجموعة . تكوين تمثيلي لـ الجسر. الـ أيون والأكسجين الأربعة
الذرات التي تنتمي إلى الاثنين المجموعات ملونة باللون البرتقالي والأخضر، على التوالي. كما نعرض جزيئات الماء الثلاثة الأيون مرتبط بتلك التكوين المحدد. ملف الطاقة الحرة بالنسبة للتفاعل بين مجموعتين كربوكسيليتين ميثيلتين، تم الإبلاغ عنه كدالة للمسافة بين أحد ذرتي الكربون من مجموعتي -CH 3 وأحد ذرتي الأكسجين المعنيتين في رابطة مزدوجة؛ (أ) التكوين النموذجي موضح في (ف)، حيث تم تلوين ذرات الكربون المعنية باللون الأرجواني. ملف الطاقة الحرة بالنسبة للتفاعل بين مجموعتين كربوكسيليتين، تم الإبلاغ عنه كدالة للمسافة بين أحد ذرتي الأكسجين المزيلتين للبروتون وواحدة من ذرتي الهيدروجين التابعتين لمجموعات الكربوكسي؛ يتم تصوير تكوين نموذجي في (h)، حيث يتم تلوين ذرات الهيدروجين المعنية باللون الليموني.
مجموعات الكربونيل، في حين أن عدد التنسيق لـ أيون مع الماء يصبح ثلاثة (على عكس القيمة 6.5 التي لوحظت في المحلول). في الواقع، في المحاكاة غير المتحيزة لربط HG التي سنناقشها لاحقًا، لم نصادف أبدًا حالة تشبه آلية “صندوق البيض”. التفاعل بين – و الوحدة قوية جدًا، كما هو موضح في ملف الطاقة الحرة المبلغ عنه في الشكل 2c. هنا، ننظر إلى مكسب الطاقة الحرة الناتج عن جلب أقرب إلى – المجموعة – بشكل خاص، المتغير الجماعي المختار هو المسافة بين وأي من ذرتي الأكسجين داخل المجموعة. يتميز الحد الأدنى من الطاقة الحرة بـ وهو قيمة تتفق بشكل ممتاز مع التقدير السابق لإمكانات القوة المتوسطة (وهو مقياس مشابه للطاقة في هذا السياق المحدد) الذي تم الإبلاغ عنه في عمل أسيفاعي وآخرون. .
على النقيض من ذلك، يبدو أن الربط المتقاطع لـ HG عبر الروابط الهيدروجينية يستفيد من تفاعلات أضعف بكثير. يتم قياس ذلك من خلال ملفات الطاقة الحرة التي حصلنا عليها بالنسبة للربط المتقاطع عبر الروابط الهيدروجينية بين إما جالورونيك أسيد الميثيل إستر و -COOH، كما هو موضح في الشكل 2e و g، على التوالي. الروابط الهيدروجينية بين مجموعات -COOH أقوى بكثير من تلك الموجودة في مجموعات الكربوكسيل الميثيلية: ضد . ومع ذلك، فإن واحداً التفاعل أقوى بكثير من أي من روابط الهيدروجين هذه. من الجدير بالذكر أن يمكن أن يحدث ذلك فقط في ظروف حمضية إلى حد ما في البداية، ومن المعقول فقط توقع أن تكون التفاعل بين اثنين من جالاكتورونيت أسيتيلated أضعف حتى من ذلك بين مجموعتين كربوكسيليتين ميثيلتين.
لذا، في ضوء هذه النتائج، نرى أن الآلية الرئيسية للتشابك المتقاطع للبيكتين منخفض الميثوكسي في الظروف الفسيولوجية للنباتات يجب أن تعتمد على – الجسور. في غياب الكفاية وعند التعامل مع البكتين عالي الميثوكسي، قد تلعب تفاعلات الروابط الهيدروجينية بين مجموعات الكربوكسيل الميثيلية دورًا مهمًا.
لكن من المؤكد أنها ستؤدي إلى بكتين مرتبط بشكل أقل قوة مقارنة بـ السيناريو. بدورنا، نؤكد أن التفاعلات الأضعف بين HG-HG ستؤدي حتماً إلى مجالات بكتين أكثر سلاسة، مما يؤدي في النهاية إلى زيادة حجم المسام المتوسطة داخل جدار الخلية.
أثر تفعيل HG على إمكانيته في الربط المتقاطع. للحصول على مزيد من الفهم حول حركيات الربط المتقاطع لـ HG، قمنا أيضًا بإجراء محاكاة غير متحيزة لتجمع سلاسل HG في الماء. على وجه الخصوص، اعتبرنا سلاسل HG تحتوي على 8 أو 40 وحدة GalA، واستكشفنا بشكل منهجي تأثير تفعيل وحدات GalA على الربط المتقاطع لـ HG. احتوت صناديق المحاكاة لدينا على ثماني سلاسل (إما بطول 8 أو 40 وحدة) تتميز بمجموعات -COOH البروتونية المختلفة. أو لسلاسل بطول 8 و 40 وحدة، على التوالي، في الجدول 1)، تم إزالة البروتون – المجموعات (D أو d في الجدول 1) ومجموعات الكربوكسيل الميثيلية ( أو في الجدول 1). نحن نلخص نتائجنا في الجدول 1، حيث نبلغ عن العدد المتوسط (المحسوب على مدار محاكاة MD التي تستمر 40 نانوثانية) من الروابط المتقاطعة (عبر جسور الكالسيوم أو عبر أي من) أو الروابط الهيدروجينية) التي تتشكل بين سلاسل HG، ومتوسط عمر هذه الروابط المتقاطعة ومتوسط عدد السلاسل المشاركة في أكبر تجمع لـ HG. لقد تحققنا من قابلية تكرار هذه النتائج عبر 15 محاكاة MD مستقلة إحصائيًا – كما تم مناقشته في المعلومات التكميلية (SI؛ انظر الجدول التكميلية 1)
نجد أن سلاسل HG التي تحتوي على D فقط تظهر أكبر مدى من الترابط المتقاطع، بينما سلاسل P التي تحتوي على P فقط ترتبط بشكل أقل فعالية، وسلاسل M التي تحتوي على M فقط ترتبط بشكل أقل من ذلك. هذه الاتجاه ليس فقط واضحًا من حيث العدد المتوسط للروابط المتقاطعة، ولكن عمر الروابط أطول بكثير لـ الجسور التي تتشكل في سلاسل D-only أكثر من الروابط الهيدروجينية التي تتشكل في سلاسل P-only أو M-only. هذه النتائج تتماشى مع الطاقة التي حصلنا عليها من خلال
الجدول 1 | الربط المتقاطع لـ HG كدالة للتوظيف
نظام سооن
بي بي بي بي بي بي بي بي 9.91(32) 1.43 6.03
دي دي دي دي دي دي دي دي 16.89 (32) 8.02 ٨
MMMMMMMM 6.06 (32) 1.65 3.94
دي دي بي دي بي 9.44 (16) 4.23 (16) 7.37 1.52 7.74
دي دي إم دي إم دي 8.66 (16) 1.71 (16) 7.88 1.53 7.35
دي بي دي بي دي بي 8.51 (16) 6.59 (16) ٨.٦ 1.62 ٧.٥١
DMDMDMDM 9.09 (16) 2.21 (16) 5.53 1.97 6.98
دي دي بي بي دي دي بي بي 8.79 (16) 4.72 (16) 8.79 1.58 7.72
دي دي إم إم دي دي إم إم 8.46 (16) 3.52 (16) 8.43 1.57 ٧.٥٤
بي بي دي دي دي دي بي بي 7.15 (16) 3.56 (16) ٥.٠٥ 1.25 ٥.٩٩
MMDDDDMM 8.16 (16) 1.71 (16) 7.11 1.33 ٤.٨١
دي دي بي بي بي بي دي دي 10.59 (16) 3.88 (16) 7.96 1.68 7.53
DDMMMMDD 9.49 (16) 2.98 (16) 6.35 2.09 7.92
DPPDDPPD 9.77 (16) 4.09 (16) 6.69 2.11 7.78
DMMDDMMD 9.77 (16) 1.75 (16) 8.62 1.37 7.38
PDDPPDDP 8.46 (16) 4.25 (16) 6.2 1.41 7.56
MDDMMDDM 7.95 (16) 5.56 (16) 8.96 1.89 7.63
دي دي دي دي بي بي بي بي 6.52 (16) 3.41 (16) 3.62 1.51 ٤.٨٦
دي دي دي دي إم إم إم إم 7.80 (16) 2.40 (16) 5.41 1.66 3.85
dddd 58.05 (160) 10.99 ٧.٩٥
مممم 1.81 (160) 1.24 2.07
dddp ٣٢.٥١ (١٢٠) 2.79 (40) 10.68 1.49 ٤.٥٢
dddm ٣٧.٩٥ (١٢٠) 0.17 (40) ١٦.٦٧ 1.11 ٥.٦١
ddpd ٣٦.١٢ (١٢٠) 0.37 (40) 12.41 1.04 6.07
ddmd ٢٩.١٤ (١٢٠) 0.98 (40) 10.67 1.66 6.54
دي دي بي بي ٣٢.٠٧ (٨٠) 8.15 (80) 11.93 1.94 5.14
يوم شهر ٢٤.٨٠ (٨٠) 0.50 (80) 14.49 1.24 ٤.٣٧
دي بي بي دي ٢٩.٤٤ (٨٠) 6.27 (80) 15.91 1.59 3.26
دي إم دي ٢٨.٨٥ (٨٠) 2.23 (80) 11.00 1.90 3.44
pddp ٢١.٢٤ (٨٠) 8.44 (80) 11.06 1.69 3.17
ميديم 16.83 (80) 0.92 (80) 9.80 1.63 2.63
دي بي دي بي ٢٦.٦٨ (٨٠) 10.37 (80) 11.43 1.87 6.22
دمدم 31.44 (80) 3.88 (80) 14.63 1.61 ٥.٣٩
دي بي بي بي 15.29 (40) 14.96 (120) 10.28 1.82 ٣.٤٥
ديمم 13.12 (40) 1.73 (120) ٢٠.١٠ 1.66 2.44
pdpp 12.24 (40) 7.92 (120) 14.53 1.45 1.93
مدم 14.46 (40) 2.04 (120) 18.8 1.36 2.00
احتوت صناديق المحاكاة لدينا على ثمانية سلاسل (إما بطول 8 وحدات أو 40 وحدة) تتميز بتوليفات مختلفة من مجموعات -COOH المؤينة (P أو p للسلاسل الطويلة 8 و40 وحدة على التوالي)، ومجموعات -COO- غير المؤينة (Dord للسلاسل الطويلة 8 و40 وحدة على التوالي) ومجموعات الكربوكسيل الميثيلية ( أو لسلاسل الطويلة التي تتكون من 8 و 40 وحدة، على التوالي). تم حساب النتائج من خلال المتوسط على محاكاة الديناميكا الجزيئية التي استمرت 40 نانوثانية. نحن نبلغ عن: العدد المتوسط للروابط المتقاطعة عبر جسور الكالسيوم، ; متوسط عدد الروابط المتقاطعة عبر روابط الهيدروجين، رؤية ; متوسط عدد الروابط المتقاطعة عبر روابط الهيدروجين -COOH، (انظر الشكل 1). كما نبلغ عن متوسط العمر الافتراضي (بالثواني النانوية) من هذه الروابط المتقاطعة بالإضافة إلى العدد المتوسط من السلاسل المشاركة في أكبر تجمع HG خلال المحاكاة، .
تمت مناقشة محاكاة الميتا ديناميكا أعلاه (وتم تلخيصها في الشكل 2). من حيث السلاسل المميزة بمزيج من وحدات D و M أو D و P، يبدو أن الكتل المكونة من 4 وحدات من وحدات مختلفة (مثل DDDDMMMM أو DDDDPPPP) تتشابك إلى حد أقل بكثير مقارنة بالسلاسل ذات التركيب نفسه حيث تكون الوحدات متناوبة على طول السلاسل (مثل DMDMDMDM أو DPDPDPDP). قد تشير هذه الاتجاهات إلى أن وحدات D المتناوبة قد تعزز من الروابط الهيدروجينية بين وحدات M أو P أيضًا. ومع ذلك، يبدو أن السلاسل المنتهية بوحدات D (أي DDPPPPDD أو DDMMMMDD) تتشابك إلى حد أكبر حتى من السلاسل المتناوبة بين D وأي من وحدات M أو P.
تتسق هذه النتائج عبر سلاسل من 8 وحدات و 40 وحدة: ومن المثير للاهتمام، أن عمر الـ الجسور في المتوسط أطول تقريبًا بمقدار الضعف مقارنةً بما حسبناه لسلاسل الـ 8 وحدات، بينما تظل مدة الحياة بالنسبة للروابط الهيدروجينية دون تغيير كبير. هذا يدل على التأثير الجماعي الناتج عن تشكيل جسور الكالسيوم المتجاورة على طول السلاسل. في الواقع، سمحت لنا محاكاة لدينا بتحديد بوضوح حدوث آلية “السحاب” التي تم مناقشتها أعلاه. في الشكل 3، نقدم لقطة تمثيلية لتجمع ثمانية سلاسل من نوع D فقط، 40 وحدة من HG. على الرغم من المدة القصيرة نسبيًا للمحاكاة، يمكننا أن نرى بوضوح تشكيل الكالسيوم.
شكل. -تداخل روابط HG بواسطة الوسائط. لقطة تمثيلية لمحاكاة MD غير متحيزة لسلاسل HG (8) ، كل منها 40 وحدة ، تم أخذها بعد تُظهر ذرات الكربون والأكسجين والهيدروجين داخل السلاسل الثلاث المعنية في الربط المتقاطع باللون السماوي والأحمر والأبيض، على التوالي. الأيونات المشاركة في الربط المتقاطع موضحة باللون البرتقالي. لاحظ ظهور آلية “السحاب” (انظر النص والشكل 1).
جسور عبر سلاسل متعددة، دون الحاجة إلى إشراك أي مجموعة – OH في العملية (كما تم الترويج له بواسطة آلية “صندوق البيض”).
بشكل عام، تشير محاكياتنا إلى أن سلاسل الهيدروجيل تتشابك مع بعضها البعض وفقًا لآلية “السحاب” التي لا تُدخل التواءً كبيرًا داخل سلسلة الهيدروجيل. بالمقابل، نلاحظ أنه لكي تشارك مجموعات -OH في الجسور الكالسيومية وفقًا لنموذج “صندوق البيض”، يجب أن تتشوه سلاسل الهيدروجيل إلى حد كبير. هذه الاعتبارات، بينما توضح التفاصيل على المستوى الجزيئي لتشابك سلاسل الهيدروجيل، تشير أيضًا إلى أن دور الرقم الهيدروجيني بالإضافة إلى محتوى الكالسيوم داخل جدار الخلية قد يكون مفتاحًا لتحديد درجة التشابك الكلي في البكتين.
لفهم تأثير الربط المتقاطع لـ HG بشكل أفضل على مسامية جدار الخلية، نتوجه إلى القياسات التجريبية التي ستناقش في القسم التالي. نلاحظ في هذه المرحلة أن عددًا كبيرًا من الجينات تتحكم في ميثيليشن البكتين (وبالتالي، لديها القدرة على التأثير على الربط المتقاطع للبكتين) وأن التكرار في وظيفتها يعني أن الطفرات الجينية تتأثر بشكل كبير بفقدان وظيفة أي جين PME واحد. على النقيض من ذلك، فإن الطفرات التي تتأثر بشكل عالمي في استرification ميثيل البكتين HG تظهر عيوب هيكلية شديدة، بما في ذلك انخفاض النمو وفقدان التصاق الخلايا، بسبب تأثيراتها على الطبقة الوسطى الغنية بـ HG. هذا يجعل من الصعب استخلاص استنتاجات حول الربط المتقاطع من خلال مقارنتها بالنباتات من النوع البري.
لتجاوز هذه المشكلة مع إجراء تغييرات ملحوظة على ترابط البكتين، استخدمنا طفرات جينية مدروسة جيدًا من أرابيدوبسيس ثاليانا التي تظهر اختلافات نوعية في مجال RGII البكتيني ولكنها تظهر كميات غير متغيرة من البكتين ولا تعاني من أي عيب في التصاق الخلايا. . هذه الطفرة تقلل من تكوين جسور البورون بين سلاسل البكتين. RG-II مرتبط تساهميًا ببكتين HG وفي النباتات من النوع البري، تعمل تلك الجسور البورونية على تقريب سلاسل البكتين من بعضها البعض وتسهيل المزيد من الربط المتقاطع من خلال الآليات التي قمنا بدراستها في محاكياتنا. . كتشكل جسور البورون وتأثيرها المتسلسل على إنشاء روابط أخرى بين سلاسل البكتين، (على سبيل المثال، HG- الجسور) متشابكة ولا يمكن فصلها تجريبيًا، نركز على فهم تأثيرها الجماعي على مسامية جدار الخلية.

التحديد التجريبي لتمسّك جدار الخلية في طافرة ربط البكتين في جدار الخلية في الأرابيدوبسيس

مؤخراً، ليو وآخرون. قدمت تقنية تعتمد على المجهر الفلوري تقيس مسامية خلايا النباتات والفطريات من خلال تقييم قدرة جزيئات كتم الفلورسنت الكبيرة نسبيًا على عبور جدار الخلية. يسمح وسم الغشاء البلازمي بالفلوريسنت بحساب درجة اختراق كاتم الفلورسنت بناءً على قدرته على كتم الفلورسنت. هذه التقنية هي تحسين أنيق للطرق المستخدمة سابقًا لتقييم حجم المسام الفعالة. لاختبار تأثير تغيير الربط المتقاطع للبكتين على مسامية جدار الخلية، طبقنا هذه التقنية، مع بعض التعديلات، على نباتات الأرابيدوبسيس من النوع البري Col0 ومتحور مدروس جيدًا، حساس للتجمد-8 (sfr8)، الذي يظهر تقليلًا في الربط المتقاطع للبكتين دون أن يظهر تشوهات تنموية كبيرة. .
كان قياس الفلورية على مدى سلسلة من النقاط الزمنية، بدلاً من قياس نقطة نهاية واحدة، مركزياً في تعديلنا للتقنية، لكشف ديناميات العملية. على وجه التحديد، قمنا بوسم الغشاء البلازمي لقشور أوراق البشرة بصبغة فلورية FM4-64 وأدخلنا جزيء كواتش، وهو الأزرق الترباني. ثم قمنا بمراقبة انخفاض الفلورية الملحوظة للعينات على مدى فترة 15 دقيقة. كما يتضح من الصور في الشكل 4a، كانت مستويات الفلورية متشابهة بين النمطين الجينيين قبل تطبيق الكواتش، لكنها انخفضت بشكل أسرع في الطفرة sfr 8 مقارنة بالنمط البري. لاحظ أن هذه الفروق كانت ذات دلالة إحصائية في لحظة التطبيق (الوقت قيمة وفي الوقت قيمة و من قيمة كما هو موضح (انظر الشكل 4ب). للتحقيق مباشرة في توزيعات حجم المسام في نباتات Col0 و sfr8 الطافرة، قمنا بإعداد عينات من أوراق Col0 و sfr8 لقياسات المجهر الإلكتروني الماسح (SEM). تم تحليل الصور (الشكل 4ج) باستخدام FIJI لتحديد كل من حجم المسام (المساحة) وتكرار كل حجم مسام (الشكل 4د). لوحظت اختلافات واضحة في توزيع حجم المسام بين النمطين الجينيين. على وجه التحديد، تشير توزيع تكرار مساحة المسام المبلغ عنها في الشكل 4د إلى أن نباتات sfr8 لديها، في المتوسط، نسبة أكبر من المسام الكبيرة (مساحة المسام ) من النوع البري Col0 ونسبة أصغر من المسام الصغيرة (مساحة المسام سمحت لنا هذه القياسات بحساب تقدير للنفاذية من العينات كـ:
أين هو متوسط حجم المسام، هو عدد المسام و هو المساحة الإجمالية للعينة. بينما كانت هناك اختلافات ملحوظة في أحجام المسام من ورقة إلى أخرى (لذا لم يكن من المناسب مقارنة أحجام المسام المتوسطة مباشرة بين النمطين الوراثيين)، نقوم بتقدير أن تكون من حيث المقدار لكلا العينتين، مع تم استخدام نتائج هذه القياسات التجريبية بعد ذلك لبناء نموذج رياضي (انظر القسم التالي) لتفسير العلاقة بين الاعتماد الزمني للفلوريسcence الملحوظة ومتوسط حجم المسام لجدار الخلية لعينات مختلفة.

النمذجة الرياضية لقياسات المسامية

للحصول على مزيد من الفهم للجوانب الميكانيكية التي تدعم النتائج التجريبية المبلغ عنها في القسم السابق، قمنا بإنشاء نموذج رياضي بسيط، استنادًا إلى اعتبارات ديناميكا السوائل. المتغيرات والمعلمات المستخدمة في هذا القسم ملخصة في الجدول 2.
كما هو موضح في الشكل 5a، نعتبر جدار خلية بمساحة وسماكة توجد جدار الخلية بين الغشاء البلازمي، مع منطقة (نفس مساحة جدار الخلية) وسمك (حيث )، وحمام من السائل الذي يحتوي على جزيء المثبط. يتم تعريف تركيز المثبط على أنه ، حيث -الإحداثي يعمل عمودياً على سطح جدار الخلية، بينما تركيز المثبط في الحمام، يُفترض أن تظل ثابتة خلال التجربة. يُفترض أن تركيز المثبط في الغشاء متجانس مكانيًا وكدالة للزمن يُعرف على أنه مع (أي، تركيز المثبط في الغشاء في بداية التجربة هو صفر). يفترض النموذج أن جدار الخلية متجانس هيكليًا (أي، لا يوجد اعتماد في على خصائص الجدار)، بحيث أن التركيز
الشكل 4 | القياسات التجريبية لتمسّك جدار الخلية في أوراق النوع البري Col0 و sfr8 من الأرابيدوبسيس. أ تغيير في الفلورية لغشاء البلازما الملون بـ FM4-64 (انظر النص) لعينات جدار الخلية من Col0 و sfr8. تم عرض سلسلة زمنية من ثلاث صور فلورية تمثيلية لكل نقطة زمنية لكل جينotype. تم التقاط صور المجهر الضوئي قبل التوقف عن التفاعل. )، مباشرة بعد إضافة صبغة التريبان الأزرق (F) وبعد ذلك كل 3 دقائق. مقياس الشريط (أصفر) هو . تغيير في الفلورية النسبية كدالة للزمن. القيم المتوسطة لـ يتم قياسها كل 3 دقائق ضمن فترة زمنية، لكل من Col0 و sfr8. القيمة عند تقاطع المحور Y تمثل العينة غير المثبطة ( ) ويعادل 1.0. تم تقييم ثلاث مناطق اهتمام (ROI) لكل تقشير بشرة (39 و 42 ROI لـ Col0 و sfr8، على التوالي). تم دمج البيانات من ثلاث تجارب منفصلة. لـ
المقارنات الإحصائية، تحليل التباين الثنائي مع اختبار المقارنات المتعددة لبونفيروني كان موظفًا. الـ القيم عند النقاط الزمنية 0 و3 و6 دقائق هي 0.0141 و0.003 و0.0221 على التوالي. لم تظهر النقاط الزمنية الأخرى أي فرق ذي دلالة إحصائية. تشير الرموز “*” و”**” إلى معتدل ( قيمة ) وقوي ( قيمة ) دليل، على التوالي، على وجود فرق إحصائي كبير بين سلسلتي الزمن. ج صور SEM لجدران خلايا Col0 و sfr8. يتم عرض ثلاث صور تمثيلية، مأخوذة من 3 أوراق مختلفة، لكلا النمطين الجينيين. شريط القياس (باللون الأصفر) هو . حجم المسام (المساحة، توزيعات التردد لـ Col0 و sfr8. تم تقييم توزيعات حجم المسام من منطقة الاهتمام (9 و 11 منطقة اهتمام لـ Col0 و sfr8، على التوالي) لثلاثة نباتات فردية لكل جينوتيب. تم استخدام توزيع تردد بعرض حاوية ونطاق قدره 15 و على التوالي، تم استخدامه.
التدرج عبره هو:
يُفترض أن جدار الخلية هو وسط مسامي، يتميز بالمسامية (أي، النسبة بين حجم الفراغات الفارغة داخل جدار الخلية والحجم الكلي لجدار الخلية) والتعقيد . الأخير هو مقياس لتعقيد المسار الذي يجب أن تسلكه السوائل (وبالتالي، في حالتنا، المثبط) للانتشار عبر الشبكة المسامية لجدار الخلية. تتميز جزيئات المثبط بمعامل انتشار في الحمام. قيم تم تقدير القيم المذكورة في الجدول 2 من خلال محاكاة الديناميكا الجزيئية لجزيء واحد من المثبط في الماء عند درجة حرارة الغرفة والضغط الجوي (انظر قسم “الطرق”). ومع ذلك، فإن معامل انتشار المثبط عبر مسام جدار الخلية من المؤكد أن يكون (أقل بكثير) من معامل الانتشار في المحلول. وبالتالي، نفترض أنه داخل المسام، يمتلك المثبط معامل انتشار ، حيث (المرجع 35). هنا، نكتب معامل الانتشار “الفعال” من المثبط في جدار الخلية كـ :
نفترض أيضًا أن المثبط يتبع الانتشار فيكي:
أين هو التدفق الانتشاري (كمية المثبط التي تنتشر في الاتجاه لكل وحدة مساحة لكل وحدة زمن. ، كمية المثبط التي تنتشر في وحدة الزمن إلى جدار الخلية هي . وبالتالي، فإن المبلغ الإجمالي الذي تم توزيعه مع مرور الوقت هو . هذا يتوزع في الغشاء، وربط ذلك بالتركيز يعطي ، وأيضًا .
من خلال دمج هذه المعادلات، نحصل على المعادلة التفاضلية العادية:
مع الحل
لديه مقياس زمني مميز
بعد أن وجدنا تركيز المثبط في الغشاء، يجب علينا الآن ربط ذلك بالفلوريسcence كما تم قياسه في التجارب. الفلوريسcence في يُشار إليه بـ ، والشدة النسبية المعرفة على أنها . هذا مرتبط بتركيز المثبط بواسطة
الجدول 2|المتغيرات والمعلمات المستخدمة في النمذجة الرياضية لقياسات المسامية
اسم رمز وحدات قيمة مرجعية
الوقت س
تركيز المثبط في الغشاء
تدفق الانتشار
شدة الفلورية
المسامية
التعقيد
معامل معدل التبريد في الغشاء ؟
مقياس زمن التحلل س ؟
ثابت المنعش
معامل انتشار المثبط في الحمام D
معامل انتشار المثبط في المسام
سمك الجدار م
سمك الغشاء م
تركيز المثبط في الحمام
منطقة العينة معروف
معادلة ستيرن-فولمر :
الذي مع (6) يعيد ترتيب إلى:
هذه هي الدالة المستخدمة لتناسب البيانات التجريبية المبلغ عنها في الشكل 5. نحن نناسب كنتيجة لـ ، مما يمنحنا قيمًا لـ و لكل نوع من النباتات.
في حالتنا، من هذا التناسب حصلنا على قيم لـ و إذا افترضنا أن المسامية (وربما التعقيد) هما الاختلافان الوحيدان بين العينتين، فإن:
لربط الالتواء بالمسامية، نستخدم تعبير بروغمان البسيط . هذا يفترض أن الهيكل المسامي متجانس ومتساوي الخواص، ومكون من جزيئات أصغر بكثير من حجم العينة. باستخدام هذا، نحصل على:
حتى:
من الملاءمة المبلغ عنها في الشكل 5 نحصل على:
الشكل 5 | النمذجة الرياضية لقياسات المسامية. أ مخططات النموذج. المتغيرات والمعلمات ذات الصلة مذكورة في الجدول 2. ب ملاءمة البيانات التجريبية (انظر الشكل 4).
تشير هذه النتيجة إلى أن جدار الخلية للطافرة sfr 8 هو أكثر مسامية من جدار الخلية من النوع البري. ومع ذلك، نؤكد أنه من أجل أن يكون هذا النموذج قادرًا على تقديم توقعات كمية عن المسامية – والأهم من ذلك – التعرج لجدار الخلية لكل عينة، هناك حاجة إلى مزيد من الجهود التجريبية، تهدف إلى، على سبيل المثال، “معايرة” النموذج من خلال قياس الاعتماد الزمني للفلوريسنس لعينات ذات مسامية و/أو تعرج معروف.
ومع ذلك، على الرغم من بساطة نموذجنا، تشير نتائجنا إلى أن: (1) يمكن تفسير النتائج التجريبية دون الحاجة إلى إدخال أي آلية “حجب” تجعل المثبط عالقًا أو تسد مسام جدار الخلية. ; (ii.) قياس تدهور الفلورية للعينة في أقرب وقت ممكن خلال التجارب مبرر من خلال حقيقة أنه من خلال القيام بذلك يتم تعظيم تأثير المسامية المختلفة للعينات؛ (iii.) المسامية المختلفة و/أو التعقيد للعينات وحدها كافية لشرح الفرق في الكثافة النسبية الملاحظة تجريبيًا. قد تستكشف الأعمال المستقبلية من منظور النمذجة نماذج أكثر تعقيدًا للتعقيد، حيث أن هذه المعلمة توفر معلومات حول شكل شبكة المسام.

نقاش

جدار الخلية النباتية هو هيكل معقد للغاية تحكمه العديد من التفاعلات بين البوليمرات الكربوهيدراتية والبروتينات الهيكلية بما في ذلك الإكستنسينات. تحدد هذه التفاعلات، ولا سيما تشكيل الروابط المتقاطعة بين سلاسل البكتين، ليس فقط القوة الفيزيائية للهيكل ولكن أيضًا مساميته، مما يؤثر على سهولة دخول مجموعة من الجزيئات الكبيرة الصغيرة المهمة بيولوجيًا، بما في ذلك الإنزيمات، وبروتينات الهيكل الخلوي، وجزيئات الإشارة بين الخلايا، والكربوهيدرات. .
لفهم أفضل لما يشكل المسامية في جدران خلايا النبات، بدأنا مع وحدات بناء البكتين وفحصنا التفاعلات الجزيئية المحتملة بين سلاسل البكتين القادرة على ربط البكتين وجعله أقل مسامية. أضافت محاكاة الديناميكا الجزيئية لدينا أدلة إلى قضية ناشئة تدعم نموذج السحاب بدلاً من نموذج صندوق البيض الشائع المفضل لدى العديد من علماء الأحياء النباتية. ومن المثير للاهتمام أن محاكاة الديناميكا الجزيئية لدينا سلطت الضوء أيضًا على أهمية الرقم الهيدروجيني على الجدار و
الربط المتقاطع. هذه البيانات تتماشى مع الملاحظات التجريبية في المختبر التي تشير إلى أن تماسك الجيلاتين من البكتين وصلابته تعتمد على كليهما و في الواقع، لقد تم تقدير حموضة الجدار منذ فترة طويلة كعامل محدد لنمو الجدار ومرونته لأسباب متنوعة، بما في ذلك اعتمادها على درجة الحموضة. رابط متقاطع لـ .
بعد التحقيق في الأساس الجزيئي لطرق تجميع سلاسل البكتين وتفاعلها مع بعضها البعض، جمعنا بيانات تجريبية من نظام بيولوجي يتم فيه تغيير الربط المتقاطع للبكتين. استخدمنا قياسات تلاشي الفلورسنت المعتمدة على الزمن لتقييم كيفية تأثير الربط المتقاطع المعدل على مسامية جدار الخلية. أكدنا صحة هذه القياسات من خلال الملاحظات المباشرة عبر المجهر الإلكتروني الماسح، ونمذجة بيانات تلاشي الفلورسنت أيضًا. في حين أن الطفرات في نبات الأرابيدوبسيس في ميثيل ترانسفيراز البكتين HG متاحة ومن المتوقع أن تظهر ربطًا متقاطعًا معدلاً، إلا أنها تعاني من العديد من العيوب بما في ذلك النمو المتعثر وفقدان التصاق الخلايا. . وبالتالي، للحصول على البيانات التجريبية لنمذجة المسامية لدينا، استخدمنا طفرة أرابيدوبسيس أكثر ملاءمة، والتي لا تظهر أي عيوب في التصاق الخلايا وتأثيرات طفيفة على النمو. حساس للتجمد-8 (sfr8) تحمل طفرة في جين الميرل المدروس جيدًا، المعروف بتأثيره على ربط البكتين RGII. . بسبب تعديل سلسلة السكر الجانبية A، يتم تقليل ترابط البكتين RGII في الطفرات sfr8/mur1. في جدران خلايا الأرابيدوبسيس من النوع البري، يمكن أن تتقاطع مجالات البكتين RGII عبر روابط ثنائي استر البورات. يتم ربط RGII بشكل تساهمي بالبكتين HG. لذا، فإن هذه الروابط تقرب سلاسل الـ HG من بعضها البعض، مما يؤدي إلى زيادة احتمالية تداخل سلاسل الـ HG مع بعضها. وهذا يتضح من الملاحظات التجريبية التي تظهر اعتماد الـ HG- الربط المتقاطع على جسور البورون RGII. كانت البكتين أكثر عرضة للاستخراج بواسطة EDTA في الطفرات murl، التي تظهر إعادة توطين غير الميثيل المستر. علاوة على ذلك، فإن وجود البورون المرتبط بـ RG-II والكالسيوم ضروريان للحفاظ على شبكة البكتين في البنية الفوقية لمخاط بذور الأرابيدوبسيس. وتؤثر على خصائص البكتين بما في ذلك اللزوجة. الفروق الملحوظة في المسامية التي لوحظت في جدران الخلايا مع وجود روابط عبر البورون في هذه المنطقة البكتينية النسبية الصغيرة سيدعم الفكرة القائلة بأن تقليل جسور البورون RGII له تأثير أكبر وأوسع انتشارًا على الروابط المتقاطعة HG أيضًا.
للتحقيق في هذه الفرضية، استخدمنا نسخة معدلة من طريقة ت quenching الفلورية التي طورها ليو وآخرون. لتحقيق في مسامية جدران خلايا الأربيذوبس البرية ونمط sfr8 المتحور. كما هو متوقع، أظهرت نتائجنا أن جزيء التهدئة اجتاز جدار الخلية لتهدئة الفلورية في الغشاء بشكل أسرع في sfr8 مقارنة بنمط Col 0 البري، مما يتماشى مع تقليل الربط المتقاطع. للتحقق من نتائجنا، استخدمنا SEM لتصوير مصفوفات جدار الخلية للأوراق وتقييم كمية المساحات المسامية المرئية في كل منها. وبما يتماشى مع نتائج الفلورية لدينا، أكد هذا أن في المتحور، تم تغيير توزيع أحجام المناطق: حيث يظهر sfr8 عددًا أقل من المناطق الصغيرة المسامية وعددًا أكبر من المناطق الكبيرة المسامية.
من أجل تنظيم قياسات الفلورية التجريبية، وخاصة لتوضيح العمليات الفيزيائية التي تدعم النتائج، قمنا ببناء نموذج رياضي بسيط يستفيد من مفاهيم ديناميكا السوائل. تشير نتائجنا إلى أن تدفق جزيئات المثبط عبر جدار الخلية لا يسد المسام داخل شبكة البكتين؛ على الرغم من أنه في المستقبل، قد تتضمن التعديلات على هذا النموذج آلية “حجب”. يمكن أن يؤخذ في الاعتبار عند التعامل مع جزيئات كواشف أكبر.
نقدم أيضًا مبررًا للاختيار التجريبي حتى الآن لقياس تدهور الفلورسنت في العينة في أقرب وقت ممكن بعد إدخال المثبط. وذلك لأن الفترات الزمنية المبكرة تعظم الفرق من حيث كفاءة التثبيط. أخيرًا، نوضح أن الاستجابات المختلفة من حيث الفلورسنت التي تثيرها عينات مختلفة يمكن تفسيرها من حيث مسامية العينات فقط.
تتيح القدرة على التنبؤ بالتفاعلات بين سلاسل البكتين في سياقات مختلفة وتوقع التغيرات في المسامية العديد من التطبيقات في تكنولوجيا الزراعة. تحدد بنية البكتين مستوى النجاح الذي قد تحققه الهجمات من الفطريات والجراثيم. وقد تم تحديد آليات حتى تستهدف الفطريات الغازية مجالات الكالسيوم-بيكتات لـ
يجعل الجدار أكثر مسامية ويسهل الوصول إلى مغذيات الخلية تشير الأدلة المتزايدة إلى أن إعادة تشكيل جدار الخلية، وخاصة هيكل البكتين، تلعب دورًا في الدفاع ضد الضغوط غير الحيوية. وهناك تقارير عن مسامية الجدار التي تؤثر على القابلية لسمية الأيونات والجفاف لقد أظهر تغيير مسامية جدار الخلية أيضًا زيادة في الكفاءة لـ امتصاص، زيادة عملية التمثيل الضوئي كنتيجة لذلك بالإضافة إلى ذلك، يمكن استغلال الفروق في مسامية جدار الخلية في تحسين امتصاص الجسيمات النانوية. جودة الخشب وقوته وتحليل المواد النباتية والطحلبية لإنتاج الوقود الحيوي هناك بوضوح فرص لحماية المحاصيل هنا.
باختصار، سلطت هذه الدراسة الضوء على الديناميات المعقدة لربط البكتين وآثاره الناتجة على مسامية جدران خلايا النبات، مقدمة رؤى هامة عبر عدة مجالات. من خلال توضيح التفاعلات الجزيئية داخل سلاسل البكتين وتأثيرها على هيكل جدار الخلية، قمنا بتعزيز فهمنا لبيولوجيا النبات، لا سيما فيما يتعلق بنمو النبات، ومقاومة الأمراض، واستجابات الإجهاد. تسلط نتائجنا الضوء على الإمكانية المتاحة لتعديل هيكل البكتين لتعزيز مرونة المحاصيل ضد مسببات الأمراض والإجهاد غير الحيوي أو تحسين إمكانيات التحلل السكري، مما يشير إلى استراتيجيات جديدة لحماية المحاصيل وتحسينها. لا تعمق هذه الأبحاث متعددة التخصصات فهمنا الأساسي لكيمياء جدران خلايا النبات فحسب، بل تفتح أيضًا آفاقًا جديدة للابتكار الزراعي والاستدامة.

طرق

نمو النباتات

تم زراعة نباتات الأرابيدوبسيس تحت 12 ساعة من الضوء: 12 ساعة من الظلام في , في غرفة نمو قابلة للدخول كما تم وصفها سابقًا . حساس للتجمد-8 (sfr8) تمت مقارنة نباتات النوع البري من كولومبيا-0 (Col0). تم زراعة النباتات حتى مرحلة أوراق الوردية (حوالي 5 أسابيع بعد الإنبات). تمت إزالة الأوراق لقياس المسامية أو التحليل بواسطة المجهر الإلكتروني الماسح.

قياسات المسامية

تم قياس الفلورية. تم تقييم مسامية جدار الخلية في قشور البشرة الورقية من خلال اختبار ت quenching الفلورية، وفقًا للطريقة التي وصفها ليو وآخرون في الأصل. لقد استخدمنا البروتوكول المفصل الموضح لأوراق الذرة في عمل ليو وآخرون. مع التعديلات التالية. تم الحصول على قشور البشرة البطنية من نباتات عمرها 5 أسابيع باستخدام تقنية فصل البشرة بالشريط المثقوب (PED) التي وصفها إيباتا وآخرون. تم تحضين القشور لمدة 5 دقائق في محلول M FM4-64 (Thermo Fisher Scientific). بعد الحضانة، تم شطف قشرة البشرة بالماء ثم تم تثبيتها على شريحة زجاجية للميكروسكوب. قبل التثبيت، تم تطبيق شريط سكوتش بعناية على طول الحواف الطويلة للشريحة الزجاجية، مما خلق غرفة في الوسط حيث تم وضع العينة وتغطيتها بشريحة تغطية. صورة أولية ( ) تم القبض عليه قبل إضافة تريبان الأزرق (سيغما-ألدريتش) لإخماد الفلورية للعينة. بعد ذلك، تم التقاط سلسلة من الصور ( ) تم الحصول عليها في فترات على مدى 15 دقيقة. تم إجراء التصوير باستخدام مجهر ليزر مسح ضوئي متماسك من زيس 800 مزود بعدسة زيتية غمر 63x ذات رقم تكبير 1.4، باستخدام ليزر تحفيز بطول موجي 488 نانومتر وكشف الفلورية عند تم استخدام برنامج ZEN Blue للتصوير الذي قدمه مصنعو المجهر (زايس) لالتقاط الصور، بينما تم إجراء التحليل باستخدام برنامج ImageJ. لتحليل الصور، متوسط شدة منطقة محددة من الاهتمام (ROI)، تحديدًا تم قياس المستطيل. تم تقييم ثلاث مناطق اهتمام منفصلة لكل عينة. تم تصدير قيم الكثافة، وتم حساب المسامية النسبية كـ .
المجهر الإلكتروني. تم الحصول على المواد الكيميائية من سيغما، بول، المملكة المتحدة أو فيشر ساينتيفيك (ما لم يُذكر خلاف ذلك). تم وضع ورقة واحدة من نبات الأرابيدوبسيس عمره 5 أسابيع في محلول PEM. ( 25 مليمول PIPES [حمض 1,4-بيبيرازيندييثان سلفونيك]، ) في درجة حرارة الغرفة وقطع إلى عرض في المحلول. عينة الأنسجة
تم نقله إلى مخزن PEM مع الفورمالديهايد (TAAB، ألدرماستون، المملكة المتحدة) ووضعه على الثلج لـ ثم تم نقل العينات إلى محلول كارنوفسكي جلوتارالدهيد (TAAB)، 2% فورمالدهيد في محلول PEM) وتم حضنه طوال الليل في . في اليوم التالي، تم غسل العينات 3-5 مرات لمدة 5-10 دقائق في محلول PEM، قبل أن تُحفظ في محلول هيبوكلوريت الصوديوم بتركيز 0.5% في محلول PEM لمدة 20 دقيقة في درجة حرارة الغرفة. بعد خمس شطفات لمدة 10 دقائق في محلول PEM، تم نقل العينات إلى أكسيد الأوزميوم الرباعي (Agar Scientific Ltd.، ستانستيد، المملكة المتحدة) في محلول PEM، لمدة 60-120 دقيقة في تم تجفيف العينات باستخدام سلسلة من الإيثانول مع في في كل من ما يلي: الإيثانول. تم الانتهاء من إزالة الماء باستخدام 3 غسلات متتالية في الإيثانول لمدة 10 دقائق في درجة حرارة الغرفة. بعد التجفيف عند نقطة حرجة (CPD) في جهاز تجفيف النقطة الحرجة ليكا EM CPD030، تم لصق العينة على قطعة من شريط لاصق مزدوج الجوانب من الكربون (قاعدة الألمنيوم؛ NISSHIN-EM CO., LTD) ملصق على حامل السيليكون (Agar Scientific). ثم تم محاط العينة بطلاء فضي (Micro to Nano، هارلم، هولندا) لتحسين التوصيلية بين العينة ورقاقة السيليكون وتم تغطيتها بطبقة رقيقة تبلغ حوالي 2 نانومتر من البلاتين بزاوية 45 درجة، باستخدام وحدة طلاء 328UHR (Cressington Scientific instruments، واتفورد، المملكة المتحدة). تم تحديد حجم المسام من أوراق تم الحصول عليها من ثلاث نباتات فردية لـ Col0 وأربع نباتات فردية لـ sfr8، باستخدام المنهجية الموضحة من قبل هوjat وآخرون. باختصار، تم معالجة صور SEM باستخدام برنامج ImageJ حيث منطقة من تم اختيارها من كل صورة للتحليل. في البداية، تم تطبيق العتبة لتقسيم الصور إلى مناطق مسامية وغير مسامية، تلاها تطبيق مرشحات شكلية، التآكل والتوسع. تم استخراج حجم وكمية المسام من كل صورة وتحويلها إلى نانومتر. كانت مناطق المسام التي تقع ضمن نطاق تم استخدامها لإنشاء رسم بياني لتوزيع التردد، بينما تم استبعاد تلك التي تقع خارج هذا النطاق من التحليل.

محاكاة الديناميكا الجزيئية

تم إجراء محاكاة الديناميكا الجزيئية باستخدام GROMACS 5.1.3 باستخدام حقل القوة CHARMM36 الشامل لجميع الذرات مع نموذج الماء TIP4P أو نموذج الماء TIP4P/Ice تم إجراء المحاكاة الأولية باستخدام TIP4P/Ice، حيث أظهر هذا النموذج فعاليته بشكل خاص في محاكاة الجزيئات الحيوية في الماء المتجمد. نظرًا لأن ديناميات هذه الأنظمة بطيئة للغاية في TIP4P/Ice، وبالتالي من الصعب استكشافها على نطاق زمن المحاكاة، فقد اخترنا الانتقال إلى نموذج الماء TIP4P، الذي يتميز عند درجة حرارة الغرفة بحركية أسرع بكثير. سمح لنا هذا الاختيار بالتحقيق في تجميع سلاسل HG دون الحاجة إلى اللجوء إلى محاكاة أخذ العينات المعززة – التي تتطلب حسابات مكلفة للغاية للتعامل مع التحقيق المنهجي الذي أجريناه هنا. يمكن العثور على تفاصيل إضافية فيما يتعلق بهذه الاستراتيجية الحسابية في الملحق، الشكل التكميلي S1.
تم إجراء محاكاة على مجموعة من الأنظمة ذات تركيبات سلاسل HG مختلفة (انظر الجدول 1). كل نظام يتكون من ثماني سلاسل متطابقة تحتوي على ثمانية بقايا لكل منها، والتي كانت إما مؤينة (P) أو غير مؤينة (D) أو ميثيل استرية (M). تحمل البقايا غير المؤينة شحنة سالبة ( شحنة، لذا يجب أن يكون هناك عدد مناسب من الكالسيوم تمت إضافة أيونات إلى الأنظمة التي تحتوي على سلاسل بها بقايا مزالة البروتونات لتحقيق توازن الشحنة الكلية. كانت البروتوكولات لهذه المحاكاة كما يلي. تم وضع سلاسل HG بشكل عشوائي في صندوق محاكاة مكعب بحجم ابتدائي 10 نانومتر. ثم تم ترطيب النظام بالماء، وإذا لزم الأمر، تم استبدال بعض جزيئات الماء بأيونات الكالسيوم لتحقيق توازن الشحنة. ثم تم إجراء تقليل الطاقة باستخدام خوارزمية الانحدار الأكثر حدة. ، تليها فترة تشغيل مدتها 60 نانوثانية، مع خطوة زمنية قدرها 2 فيمتوثانية، عند درجة حرارة الغرفة والضغط المحيط في مجموعة NPT. منظم الحرارة بوسي-دوناديو-بارينيلو وبارست بريندسن تم استخدام قيم اقتران تبلغ 0.5 و 4 بيكوثانية، على التوالي. تم تطبيق شروط الحدود الدورية في ثلاثة أبعاد. يمكن العثور على تفاصيل حسابية إضافية تتعلق بهذه المحاكاة الديناميكية الجزيئية في الملحق، الجدول التكميلي 2.
للبحث بالتفصيل في طاقة التفاعلات المختلفة الموضحة في الشكل 1، لجأنا إلى الميتاديناميكا المدروسة.
محاكاة باستخدام PLUMED 2.4.2 . الديناميكا الميتا المنضبطة هي تقنية تحسين العينة التي تسمح باستكشاف سطح الطاقة الحرة للنظام بالنسبة لمجموعة مختارة من درجات الحرية، أو “المتغيرات الجماعية” (CVs). تم استخدام هذه المحاكاة لقياس القوة النسبية لطريقتي الربط HG، وهما الجسور الكالسيومية والروابط الهيدروجينية. احتوت المحاكاة على سلسلتين متطابقتين من HG مكونتين من ثمانية بقايا، مع سبع بقايا بروتونية وبقايا واحدة إما غير بروتونية أو ميثيل استرية. بالنسبة للجسر الكالسيومي، تم اختيار المسافة بين أيون الكالسيوم والموقع غير البروتوني كمتغير جماعي للتأثير: تم ضبط عرض وارتفاع وخطوة الإيداع للجهود الغاوسية (بعد اختبارات شاملة وتحقق) إلى ، و500 خطوة، على التوالي. تم تعيين عامل التحيز إلى . بالنسبة للرابطة الهيدروجينية، تم اختيار المسافة بين كربون الميثيل في أحد المتبقيات وأكسجين الكربونيل في المتبقي الآخر كمتغير تحكم للتأثير: تم ضبط العرض والارتفاع وخطوة الإيداع لإمكانات غاوسي (بعد اختبارات شاملة وتحقق) إلى ، و500 خطوة، على التوالي. تم تعيين عامل التحيز إلى يمكن العثور على تفاصيل حسابية إضافية تتعلق بمحاكاة الميتا ديناميكس في الملحق، الجدول التكميلي 3.
تم تعيين معدل ترسيب الانحياز إلى 500 خطوة MD (ما يعادل 1 بيكوثانية) حيث أن زمن الاسترخاء الهيكلي لماء TIP4P عند درجة حرارة الغرفة والضغط هو من حيث تم تعيين عرض الإمكانيات الغاوسية في البداية إلى . تبين أن هذه القيمة صغيرة جدًا، حيث أن الأسطح الناتجة للطاقة الحرة واجهت صعوبة في تحقيق التقارب بسبب الدقة العالية جدًا. لذلك، قمنا بزيادة إلى 0.04 و 0.05 نانومتر للتحقيق في تفاعلات جسر الكالسيوم وروابط الهيدروجين، على التوالي. قيم نتج عن ذلك أن أسطح الطاقة الحرة تفتقر إلى التفاصيل/الدقة الكافية. تم تحديده في البداية لنصف قيمة الطاقة الحرارية ( ) في درجة حرارة الغرفة، أي، . هذه قاعدة شائعة من حيث التوصل إلى تسوية بين كفاءة العينة والدقة. لقد استكشفنا قيم بين 1.00 و ، بالتزامن مع عوامل التحيز (التي تؤثر قيمتها أيضًا على نسبة الكفاءة/ الدقة) التي تتراوح من 50 إلى 100. أي مجموعة من قيم و ضمن النطاقات المذكورة أعلاه، أدى بنا إلى الحصول على أسطح طاقة حرة ضمن عدم اليقين المحسوب من خلال تقنية إعادة الوزن لتيواري وبارينيلو. (وتم الإبلاغ عنها في الشكل 2 كمناطق زرقاء مظللة) ضمن وقت محاكاة مقبول ( ).
بينما لا يحمل مفهوم الزمن في محاكاة الميتا ديناميكا معنى فيزيائي مباشر (حيث أن تطور الزمن للنظام يتأثر بشدة بالتحيز الاصطناعي المقدم لأغراض استكشاف سطح الطاقة الحرة المعني)، كانت هذه الأوقات الطويلة للمحاكاة ضرورية لضمان التقارب الصحيح لسطح الطاقة الحرة الناتج. لتحقيق ذلك، قمنا بـ: (i.) مراقبة التغيرات في سطح الطاقة الحرة كدالة لزمن المحاكاة؛ (ii.) استكشاف تأثير تغيير ارتفاع وعرض جهد التحيز؛ (iii.) استخدام تقنية إعادة الوزن لتيواري وبارينيلو. لتقديم تقدير كمي لعدم اليقين المرتبط بسطح الطاقة الحرة.
الإحصائيات وإمكانية التكرار. لقد قمنا بتضمين مناقشة حول إمكانية تكرار نتائجنا فيما يتعلق بكل جانب من جوانب منهجيتنا (انظر قسم الطرق). ربما يكون الجانب الأكثر حساسية في عملنا يتعلق بإمكانية تكرار نتائجنا من حيث محاكاة الديناميات الجزيئية لتجمعات HG (انظر الجدول 1). لهذا الغرض، قمنا بتضمين دليل مباشر على قوة هذه النتائج في المعلومات التكميلية. فيما يتعلق بقياسات الفلورية المبلغ عنها في الشكل 4ب، تم تقييم ثلاث مناطق منفصلة من الاهتمام (ROI) لكل قشرة جلدية (بمجموع 39 و42 ROI لـ Col 0 و sfr8، على التوالي). تم دمج البيانات من ثلاث تجارب منفصلة. للمقارنات الإحصائية، تم استخدام تحليل التباين ثنائي الاتجاه مع اختبار المقارنات المتعددة لبونفيروني. تم التوظيف. تم تحديد الأهمية عند . من حيث قياسات SEM، تم تقييم توزيعات حجم المسام من منطقة الاهتمام (9 و 11 منطقة اهتمام لـ Col0 و sfr8، على التوالي) لثلاثة نباتات فردية لكل نوع وراثي. تم استخدام توزيع تكراري بعرض حاوية ونطاق قدره 15 و تم استخدام النسخة 10.1.0 من برنامج GraphPad Prism لنظام التشغيل ويندوز. فيما يتعلق بنتائج الديناميات الميتا
في المحاكاة المبلغ عنها في الشكل 2، استخدمنا تقنية إعادة الوزن التي اقترحها تيوري وبارينيلو لتقييم عدم اليقين المرتبط بملفات الطاقة الحرة الناتجة.

ملخص التقرير

معلومات إضافية حول تصميم البحث متاحة في ملخص تقارير مجموعة ناتشر المرتبط بهذه المقالة.

توفر البيانات

جميع البيانات المرتبطة بهذه الدراسة متاحة من خلال مستودع GitHub عام الوصول (PEC_LINK،https://github.com/gcsosso/PEC_LINK. git). لقد ربطنا أيضًا إصدارًا من هذا المستودع بمستودع Zenodo (المتاح للجمهور أيضًا)، والذي حصلنا من خلاله على DOI دائم. يمكن الحصول على أي معلومات متبقية من المؤلف المقابل عند الطلب المعقول.

توفر الشيفرة

يمكن العثور على تفاصيل الأكواد المستخدمة في هذا العمل في قسم الطرق. لم يتم استخدام أي كود داخلي أو ملكي. فيما يتعلق بالجوانب الحاسوبية للعمل، نلاحظ أن كل من حزمتي GROMACS و PLUMED مفتوحة المصدر ومتاحة للجمهور.
تاريخ الاستلام: 8 أبريل 2024؛ تاريخ القبول: 8 يناير 2025؛
نُشر على الإنترنت: 17 يناير 2025

References

  1. Cosgrove, D. J. Growth of the plant cell wall. Nat. Rev. Mol. Cell Biol. 6, 850-861 (2005).
  2. Delmer, D., Dixon, R.A., Keegstra, K., Mohnen, D. The plant cell walldynamic, strong, and adaptable-is a natural shapeshifter. Plant Cell, 325. https://doi.org/10.1093/plcell/koad325 (2024).
  3. Caffall, K. H. & Mohnen, D. The structure, function, and biosynthesis of plant cell wall pectic polysaccharides. Carbohydr. Res. 344, 1879-1900 (2009).
  4. Baron-Epel, O., Gharyal, P. K. & Schindler, M. Pectins as mediators of wall porosity in soybean cells. Planta 175, 389-395 (1988).
  5. Fleischer, A., O’Neill, M. A. & Ehwald, R. The pore size of nongraminaceous plant cell walls is rapidly decreased by borate ester cross-linking of the pectic polysaccharide rhamnogalacturonan II1. Plant Physiol. 121, 829-838 (1999).
  6. Bidhendi, A. J. & Geitmann, A. Relating the mechanics of the primary plant cell wall to morphogenesis. J. Exp. Bot. 67, 449-461 (2016).
  7. Carpita, N., Sabularse, D., Montezinos, D. & Delmer, D. P. Determination of the pore size of cell walls of living plant cells. Science 205, 1144-1147 (1979).
  8. Pérez-Donoso, A. G., Sun, Q., Roper, M. C., Greve, L. C., Kirkpatrick, B. & Labavitch, J. M. Cell wall-degrading enzymes enlarge the pore size of intervessel pit membranes in healthy and Xylella fastidiosainfected grapevines. Plant Physiol. 152, 1748-1759 (2010).
  9. Savary, S., Willocquet, L., Pethybridge, S. J., Esker, P., McRoberts, N. & Nelson, A. The global burden of pathogens and pests on major food crops. Nat. Ecol. Evol. 3, 430-439 (2019).
  10. Adani, F., Papa, G., Schievano, A., Cardinale, G., D’Imporzano, G. & Tambone, F. Nanoscale structure of the cell wall protecting cellulose from enzyme attack. Environ. Sci. Technol. 45, 1107-1113 (2011).
  11. Wang, Y., Xu, X., Xue, H., Zhang, D. & Li, G. Physical-chemical properties of cell wall interface significantly correlated to the complex recalcitrance of corn straw. Biotechnol. Biofuels 14, 196 (2021).
  12. Grant, G. T., Morris, E. R., Rees, D. A., Smith, P. J. C. & Thom, D. Biological interactions between polysaccharides and divalent cations: The egg-box model. FEBS Lett. 32, 195-198 (1973).
  13. Zdunek, A., Pieczywek, P. M. & Cybulska, J. The primary, secondary, and structures of higher levels of pectin polysaccharides. Compr. Rev. Food Sci. Food Saf. 20, 1101-1117 (2021).
  14. Assifaoui, A. et al. Structural behaviour differences in low methoxy pectin solutions in the presence of divalent cations (Ca2+ and Zn2+): a process driven by the binding mechanism of the cation with the galacturonate unit. Soft Matter 11, 551-560 (2014).
  15. Poset, A. M. D., Zitolo, A., Cousin, F., Assifaoui, A. & Lerbret, A. Evidence for an egg-box-like structure in iron(II)-polygalacturonate hydrogels: a combined EXAFS and molecular dynamics simulation study. Phys. Chem. Chem. Phys. 22, 2963-2977 (2020).
  16. Pelloux, J., Rustérucci, C. & Mellerowicz, E. J. New insights into pectin methylesterase structure and function. Trends Plant Sci. 12, 267-277 (2007).
  17. Harholt, J., Suttangkakul, A. & Vibe Scheller, H. Biosynthesis of pectin. Plant Physiol. 153, 384-395 (2010).
  18. Wormit, A. & Usadel, B. The multifaceted role of pectin methylesterase inhibitors (PMEls). Int J. Mol. Sci. 19, 2878 (2018).
  19. O’Neill, M. A., Ishii, T., Albersheim, P. & Darvill, A. G. RHAMNOGALACTURONAN II: structure and function of a borate cross-linked cell wall pectic polysaccharide.Annu. Rev. Plant Biol. 55, 109-139 (2004).
  20. Rondeau-Mouro, C., Defer, D., Leboeuf, E. & Lahaye, M. Assessment of cell wall porosity in Arabidopsis thaliana by NMR spectroscopy. Int. J. Biol. Macromol. 42, 83-92 (2008).
  21. Liu, X. et al. Novel tool to quantify cell wall porosity relates wall structure to cell growth and drug uptake. J. Cell Biol. 218, 1408-1421 (2019).
  22. Fu, J., Thomas, H. R. & Li, C. Tortuosity of porous media: Image analysis and physical simulation. Earth Sci. Rev. 212, 103439 (2021).
  23. Du, J., Anderson, C. T. & Xiao, C. Dynamics of pectic homogalacturonan in cellular morphogenesis and adhesion, wall integrity sensing and plant development. Nat. Plants 8, 332-340 (2022).
  24. Plazinski, W. Molecular basis of calcium binding by polyguluronate chains. Revising the egg-box model. J. Comput Chem. 32, 2988-2995 (2011).
  25. Temple, H. et al. Golgi-localized putative S-adenosyl methionine transporters required for plant cell wall polysaccharide methylation. Nat. Plants 8, 656-669 (2022).
  26. Draget, K. I., Stokke, B. T., Yuguchi, Y., Urakawa, H. & Kajiwara, K. Small-angle X-ray scattering and rheological characterization of alginate gels. 3. alginic acid gels. Biomacromolecules 4, 1661-1668 (2003).
  27. Ventura, I., Jammal, J. & Bianco-Peled, H. Insights into the nanostructure of low-methoxyl pectin-calcium gels. Carbohydr. Polym. 97, 650-658 (2013).
  28. Shi, D.-C. et al. Boron-bridged RG-II and calcium are required to maintain the pectin network of the Arabidopsis seed mucilage ultrastructure. Plant Mol. Biol. 94, 267-280 (2017).
  29. Braccini, I. & Pérez, S. Molecular basis of Ca2+-induced gelation in alginates and pectins: the egg-box model revisited. Biomacromolecules 2, 1089-1096 (2001).
  30. Gilsenan, P. M., Richardson, R. K. & Morris, E. R. Thermally reversible acid-induced gelation of low-methoxy pectin. Carbohydr. Polym. 41, 339-349 (2000).
  31. Yoo, J., Wilson, J. & Aksimentiev, A. Improved model of hydrated calcium ion for molecular dynamics simulations using classical biomolecular force fields. Biopolymers 105, 752-763 (2016).
  32. Panter, P. E. et al. Cell wall fucosylation in Arabidopsis influences control of leaf water loss and alters stomatal development and mechanical properties. J. Exp. Bot. 74, 2680 (2023).
  33. Ishii, T. & Matsunaga, T. Pectic polysaccharide rhamnogalacturonan II is covalently linked to homogalacturonan. Phytochemistry 57, 969-974 (2001).
  34. Panter, P. E. et al. MUR1-mediated cell-wall fucosylation is required for freezing tolerance in Arabidopsis thaliana. N. Phytol. 224, 1518-1531 (2019).
  35. Bruna, M. & Chapman, S. J. Diffusion of finite-size particles in confined geometries. Bull. Math. Biol. 76, 947-982 (2014).
  36. Tartakovsky, D. M. & Dentz, M. Diffusion in porous media: phenomena and mechanisms. Transp. Porous Media 130, 105-127 (2019).
  37. Lakowicz, J.R. (ed). Principles of Fluorescence Spectroscopy. https:// doi.org/10.1007/978-0-387-46312-4 (Springer, 2006).
  38. Tjaden, B., Cooper, S. J., Brett, D. J., Kramer, D. & Shearing, P. R. On the origin and application of the Bruggeman correlation for analysing transport phenomena in electrochemical systems. Curr. Opin. Chem. Eng. 12, 44-51 (2016).
  39. Herterich, J. G., Griffiths, I. M. & Vella, D. Reproducing the pressure-time signature of membrane filtration: The interplay between fouling, caking, and elasticity. J. Membr. Sci. 577, 235-248 (2019).
  40. Castilleux, R., Plancot, B., Vicré, M., Nguema-Ona, E. & Driouich, A. Extensin, an underestimated key component of cell wall defence? Ann. Bot. 127, 709-713 (2021).
  41. Read, S.M., Bacic, A. Cell wall porosity and its determination. in Plant Cell Wall Analysis (eds Linskens, H.F., & Jackson, J.F.) 63-80. https:// doi.org/10.1007/978-3-642-60989-3_4 (Springer, 1996).
  42. Löfgren, C., Guillotin, S., Evenbratt, H., Schols, H. & Hermansson, A.M. Effects of calcium, pH, and blockiness on kinetic rheological behavior and microstructure of HM pectin gels. Biomacromolecules 6, 646-652 (2005).
  43. Yang, X. et al. Low methoxyl pectin gelation under alkaline conditions and its rheological properties: Using NaOH as a pH regulator. Food Hydrocoll. 79, 560-571 (2018).
  44. Phyo, P., Gu, Y. & Hong, M. Impact of acidic pH on plant cell wall polysaccharide structure and dynamics: insights into the mechanism of acid growth in plants from solid-state NMR. Cellulose 26, 291-304 (2019).
  45. Mouille, G. et al. Homogalacturonan synthesis in Arabidopsis thaliana requires a golgi-localized protein with a putative methyltransferase domain. Plant J. 50, 605-614 (2007).
  46. Kim, S.-J., Held, M. A., Zemelis, S., Wilkerson, C. & Brandizzi, F. CGR2 and CGR3 have critical overlapping roles in pectin methylesterification and plant growth in Arabidopsis thaliana. Plant J. 82, 208-220 (2015).
  47. Du, J. et al. Mutations in the pectin methyltransferase QUASIMODO2 influence cellulose biosynthesis and wall integrity in Arabidopsis. Plant Cell 32, 3576-3597 (2020).
  48. O’Neill, M. A., Eberhard, S., Albersheim, P. & Darvill, A. G. Requirement of borate cross-linking of cell wall rhamnogalacturonan II for Arabidopsis growth. Science 294, 846-849 (2001).
  49. Bellincampi, D., Cervone, F., Lionetti, V. Plant cell wall dynamics and wall-related susceptibility in plant-pathogen interactions. Front. Plant Sci. 5, 228 (2014).
  50. Forand, A. D. et al. With a little help from my cell wall: structural modifications in pectin may play a role to overcome both dehydration stress and fungal pathogens. Plants 11, 385 (2022).
  51. Vogel, J. P., Raab, T. K., Schiff, C. & Somerville, S. C. PMR6, a pectate lyase-like gene required for powdery mildew susceptibility in Arabidopsis. Plant Cell 14, 2095-2106 (2002).
  52. Tenhaken, R. Cell wall remodeling under abiotic stress. Front. Plant Sci. 5, 771 (2015).
  53. Panter, P.E., Panter, J.R., Knight, H. Impact of cell-wall structure and composition on plant freezing tolerance. In Annual Plant Reviews Online 607-642. https://doi.org/10.1002/9781119312994.apr0746 (John Wiley & Sons, 2020).
  54. Wu, H.-C., Bulgakov, V. P. & Jinn, T.-L. Pectin methylesterases: cell wall remodeling proteins are required for plant response to heat stress. Front. Plant Sci. 9, 1612 (2018).
  55. Yang, Z.-B., Eticha, D., Rao, I. M. & Horst, W. J. Alteration of cell-wall porosity is involved in osmotic stress-induced enhancement of
    aluminium resistance in common bean (Phaseolus vulgaris L.). J. Exp. Bot. 61, 3245 (2010).
  56. Dang, Z. et al. The Fragile culm19 (FC19) mutation largely improves plant lodging resistance, biomass saccharification, and cadmium resistance by remodeling cell walls in rice. J. Hazard. Mater. 458, 132020 (2023).
  57. Salesse-Smith, C. E. et al. Greater mesophyll conductance and leaf photosynthesis in the field through modified cell wall porosity and thickness via AtCGR3 expression in tobacco. Plant Biotechnol. J. 22, 2504-2517 (2024).
  58. Wu, H. & Li, Z. Nano-enabled agriculture: How do nanoparticles cross barriers in plants? Plant Commun. 3, 100346 (2022).
  59. Li, X. & Zhao, Z. Time domain-NMR studies of average pore size of wood cell walls during drying and moisture adsorption. Wood Sci. Technol. 54, 1241-1251 (2020).
  60. Loqué, D., Scheller, H. V. & Pauly, M. Engineering of plant cell walls for enhanced biofuel production. Curr. Opin. Plant Biol. 25, 151-161 (2015).
  61. Wang, Y., Xu, X., Xue, H., Zhang, D., Li, G. Recalcitrance of corn straw significantly correlated to the heterogeneity in physicochemical properties of cell wall. Res. Sq. https://doi.org/10.21203/rs.3.rs-605820/v1 (2021).
  62. Liu, X., Pomorski, T. G. & Liesche, J. Non-invasive quantification of cell wall porosity by fluorescence quenching microscopy. Bio. Protoc. 9, 3344 (2019).
  63. Ibata, H., Nagatani, A. & Mochizuki, N. Perforated-tape epidermal detachment (PED): a simple and rapid method for isolating epidermal peels from specific areas of Arabidopsis leaves. Plant Biotechnol. 30, 497-502 (2013).
  64. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676-682 (2012).
  65. Sugimoto, K., Williamson, R. E. & Wasteneys, G. O. New techniques enable comparative analysis of microtubule orientation, wall texture, and growth rate in intact roots of Arabidopsis. Plant Physiol. 124, 1493-1506 (2000).
  66. Hojat, N., Gentile, P., Ferreira, A. M. & Šiller, L. Automatic pore size measurements from scanning electron microscopy images of porous scaffolds. J. Porous Mater. 30, 93-101 (2023).
  67. Abraham, M. J. et al. Gromacs: high performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1-2, 19-25 (2015).
  68. Guvench, O. et al. CHARMM additive all-atom force field for carbohydrate derivatives and its utility in polysaccharide and carbohydrate- protein modeling. J. Chem. Theory Comput. 7, 3162-3180 (2011).
  69. Lawrence, C. P. & Skinner, J. L. Flexible TIP4P model for molecular dynamics simulation of liquid water. Chem. Phys. Lett. 372, 842-847 (2003).
  70. Abascal, J.L.F., Sanz, E., Fernández, R.G., Vega, C. A potential model for the study of ices and amorphous water: TIP4P/Ice. J. Chem. Phys. 122, 234511 (2005).
  71. Lee, H. Interactions of antifreeze proteins in TIP4P / Ice water and their dependence on force fields. PLoS ONE 07, 1-17 (2018).
  72. Mochizuki, K. & Molinero, V. Antifreeze glycoproteins bind reversibly to ice via hydrophobic groups. J. Am. Chem. Soc. 140, 4803-4811 (2018).
  73. Sosso, C. G. et al. Unravelling the origins of ice nucleation on organic crystals. Chem. Sci. 9, 8077-8088 (2018).
  74. Mayne, D. Q. A stochastic steepest-descent algorithm. J. Optim. Theory Appl. 59, 307-323 (1988).
  75. Bussi, G., Donadio, D. & Parrinello, M. Canonical sampling through velocity rescaling. J. Chem. Phys. 126, 1-7 (2007).
  76. Berendsen, H. J. C., Postma, J. P. M., van Gunsteren, W. F., DiNola, A. & Haak, J. R. Molecular dynamics with coupling to an external bath. J. Chem. Phys. 81, 3684-3690 (1984).
  77. Laio, A. & Parrinello, M. Escpaing free-energy minima. PNAS 99, 12562-12566 (2002).
  78. Barducci, A. & Bussi, G. Well-tempered metadynamics: a smoothly converging and tunable free- energy method. Phys. Rev. Lett. 100, 1-4 (2008).
  79. Tribello, G. A., Bonomi, M., Branduardi, D., Camilloni, C. & Bussi, G. PLUMED 2: new feathers for an old bird. Comput. Phys. Commun. 185, 604-613 (2014).
  80. Gallo, P. & Rovere, M. Relation between the two-body entropy and the relaxation time in supercooled water. Phys. Rev. E 91, 012107 (2015).
  81. Tiwary, P. & Parrinello, M. A time-independent free energy estimator for metadynamics. J. Phys. Chem. B 119, 736-742 (2015).
  82. McHugh, M. L. Multiple comparison analysis testing in ANOVA. Biochem. Med. 21, 203-209 (2011).
  83. Sosso, G.C. PEC_LINK. Accessed: 2025-01-08 (2024). https:// zenodo.org/records/14366068

شكر وتقدير

تم دعم هذا العمل من قبل مجلس بحوث العلوم البيولوجية والتكنولوجيا الحيوية (BBSRC) منحة “فهم تكوين الجليد في النباتات: إيجاد طرق جديدة لتحمل التجمد (Plantlce)” (BB/N015559/1). تم تمويل N.R. من خلال منحة تدريب الدكتوراه من BBSRC (BB/M011186/1). يتم تمويل P.L.J. عبر مركز تدريب EPSRC في نمذجة الأنظمة غير المتجانسة (HetSys، رقم المنحة EP/S022848/1). يقر J.E.S بالدعم من EPSRC بموجب المنح EP/W031426/1 و EP/S029966/1 و EP/P031684/1. يقر I.J.P. و F.B. و G.C.S. بامتنان باستخدام خدمة الحوسبة الفائقة الوطنية ARCHER2 في المملكة المتحدة.https://www.archer2.ac.uk)، والتي تم الوصول إليها عبر اتحاد HecBioSim (المشروع e676)، الممول من قبل EPSRC (رقم المنحة EP/X035603/1). كما يود I.J.P. و G.C.S. أن يعربا عن امتنانهما لاستخدام SULIS، الذي تم تمويله من قبل EPSRC (EP/T022108/1)، عبر اتحاد HPC Midlands+. كما يود I.P. و F.B. و G.C.S. أن يعبروا عن شكرهم لمرافق الحوسبة عالية الأداء المقدمة من منصة تكنولوجيا البحث في الحوسبة العلمية (SCRTP) في جامعة ووريك. كما يود H.K و T.J.H و I.O. أن يعربوا عن شكرهم لأندرو إيسكاوسكاس على المناقشات الحاسمة/ المفيدة حول التحليل الإحصائي لقياسات SEM.

مساهمات المؤلفين

قام كل من ج.س.، هـ.ك، ت.ج.هـ، إ.ع. و إ.ج.ب. بتصور البحث. قام إ.ج.ب، ف.ب. و ج.س. بتنفيذ وتحليل محاكاة الديناميكا الجزيئية والديناميكا الميتا. قام ب.ل.ج، ج.س. و ج.إ.س. بتصميم وتطبيق وتحليل النموذج الرياضي القائم على ديناميكا السوائل. قام إ.ع.، ت.ج.هـ، هـ.ك. و م.و.ج. بتصميم الدراسات التجريبية. قام إ.ع. و ن.ر. بتوسيع المنهجية و
تم إجراء اختبارات المسامية المعتمدة على الفلورية. قام C.K.I. بتطوير منهجية SEM وأجرى، مع I.O.، قياسات EM. ساهم I.O. وI.J.P. وP.L.J. وF.B. وT.J.H. وJ.E.S. وH.K. وG.C.S. في كتابة المخطوطة.

المصالح المتنافسة

يعلن المؤلفون عدم وجود مصالح متنافسة.

معلومات إضافية

معلومات إضافية النسخة الإلكترونية تحتوي على مواد إضافية متاحة في
https://doi.org/10.1038/s42003-025-07495-0.
يجب توجيه المراسلات والطلبات للحصول على المواد إلى هيذر نايت أو غابرييل سي. سوسو.
تشكر مجلة Communications Biology دوودو وان والمراجعين الآخرين المجهولين على مساهمتهم في مراجعة الأقران لهذا العمل. المحررون الرئيسيون: كاليا جورجيفا. يتوفر ملف مراجعة الأقران.
معلومات إعادة الطبع والتصاريح متاحة علىhttp://www.nature.com/reprints
ملاحظة الناشر: تظل Springer Nature محايدة فيما يتعلق بالمطالبات القضائية في الخرائط المنشورة والانتماءات المؤسسية.
الوصول المفتوح: هذه المقالة مرخصة بموجب رخصة المشاع الإبداعي للاستخدام الدولي 4.0، والتي تسمح بالاستخدام والمشاركة والتكيف والتوزيع وإعادة الإنتاج بأي وسيلة أو صيغة، طالما أنك تعطي الائتمان المناسب للمؤلفين الأصليين والمصدر، وتوفر رابطًا لرخصة المشاع الإبداعي، وتوضح ما إذا تم إجراء تغييرات. الصور أو المواد الأخرى من طرف ثالث في هذه المقالة مشمولة في رخصة المشاع الإبداعي للمقالة، ما لم يُذكر خلاف ذلك في سطر الائتمان للمادة. إذا لم تكن المادة مشمولة في رخصة المشاع الإبداعي للمقالة وكان استخدامك المقصود غير مسموح به بموجب اللوائح القانونية أو يتجاوز الاستخدام المسموح به، ستحتاج إلى الحصول على إذن مباشرة من صاحب حقوق الطبع والنشر. لعرض نسخة من هذه الرخصة، قم بزيارةhttp://creativecommons.org/licenses/by/4.0/.
© المؤلفون 2025

  1. ¹قسم علوم الحياة ومركز ديرم لتحسين المحاصيل، جامعة ديرم، ديرم، المملكة المتحدة. ²قسم الكيمياء، جامعة ووريك، كوفنتري، المملكة المتحدة. معهد الرياضيات بجامعة ووريك، كوفنتري، المملكة المتحدة. □ البريد الإلكتروني:p.h.knight@durham.ac.uk;g.sosso@warwick.ac.uk

Journal: Communications Biology, Volume: 8, Issue: 1
DOI: https://doi.org/10.1038/s42003-025-07495-0
PMID: https://pubmed.ncbi.nlm.nih.gov/39825091
Publication Date: 2025-01-17

Understanding pectin cross-linking in plant cell walls

Irabonosi Obomighie (1) , lain J. Prentice , Peter Lewin-Jones () , Fabienne Bachtiger , Nathan Ramsay , Chieko Kishi-Itakura , Martin W. Goldberg , Tim J. Hawkins , James E. Sprittles , Heather Knight & Gabriele C. Sosso
Pectin is a major component of plant cells walls. The extent to which pectin chains crosslink with one another determines crucial properties including cell wall strength, porosity, and the ability of small, biologically significant molecules to access the cell. Despite its importance, significant gaps remain in our comprehension, at the molecular level, of how pectin cross-links influence the mechanical and physical properties of cell walls. This study employs a multidisciplinary approach, combining molecular dynamics simulations, experimental investigations, and mathematical modelling, to elucidate the mechanism of pectin cross-linking and its effect on cell wall porosity. The computational aspects of this work challenge the prevailing egg-box model, favoring instead a zipper model for pectin cross-linking, whilst our experimental work highlights the significant impact of pectin cross-linking on cell wall porosity. This work advances our fundamental understanding of the biochemistry underpinning the structure and function of the plant cell wall. This knowledge has important implications for agricultural biotechnology, informing us about the chemical properties of plant pectins that are best suited for improving crop resilience and amenability to biofuel extraction by modifying the cell wall.
Unlike animal cells, plant cells are surrounded by a cell wall, which provides strength and protection, whilst allowing growth . As well as its structural role, the cell wall constitutes an important barrier that determines what can and cannot gain access to the cell. It is porous, allowing the passage of water and gases but the degree of porosity may vary, and this is what determines which small molecules are admitted to the cell or impeded from entry. The primary cell wall of plants is composed of complex carbohydrates including pectins, cellulose and hemicelluloses and embedded proteins with structural and enzymatic functions . Pectins are characterised by a galacturonic acid (GalA)-rich backbone and form covalent links with each other , resulting in the formation of the complex 3-D structure of the cell wall that governs many properties, including mechanical strength and wall porosity . The first estimates of cell wall porosity demonstrated that the wall provided a more significant barrier to the passage of small molecules than was previously assumed, suggesting that control of porosity could regulate cellular communication . Nowadays, understanding and predicting plant cell wall pore size is critical to plant biotechnology and agriculture. Small cell wall pore sizes can restrict the entry and efficacy of cell wall-degrading enzymes, such as cellulases, used by invasive pathogens , which cause phenomenal crop
losses worldwide . Conversely, for the same reason, porous walls are more amenable to biofuel production by the process of saccharification .
Pectin gelation is affected by cross-links forming between pectin chains or their sugar side chains. Homogalacturonan (HG) pectins, which represent the majority of cell-wall pectin, primarily cross-link to one another via calcium ions ( ) between demethylesterified (DM) carboxyl groups, as illustrated in Fig. 1. This cross-linking mechanism is thought to lead to the aggregation of HG chains via the so-called “egg-box” model , although recent work has put the validity of this model into question. The degree of pectin DM is determined by the action of a large family of enzymes, the pectin methylesterases (PMEs) and by numerous PME inhibitors (PMEIs) . Demethylesterfied HG can also link to the other pectic domains and xyloglucans . The more complex pectic domain rhamnogalacturonan II (RGII) chains cross-link to one another via borate diester linkages between adjacent sugar side chains . In each case, the degree of cross-linking affects porosity, i.e., the nanometre to micrometre scale spaces within the wall .
In this work, we have leveraged a multidisciplinary approach to shed light onto the mechanism of cross-linking of pectin and to establish a connection between the structure of pectin and the porosity of the cell wall.
Fig. 1 | Different cross-linking mechanisms for HG chains. The different functional groups involved, namely protonated carboxyl groups (-COOH), deprotonated carboxyl groups (- ), methylated carboxyl groups (methyl-esterified GalA) and acetylated carboxyl groups, are highlighted in purple, green, orange and pink, respectively. Two distinct mechanisms have been proposed for the crosslinking of HG chains via bridges: the “egg-box” and the “zipper” model. The
egg-box model is commonly found in the literature revolving around pectin crosslinking, and involves two hydroxyl (- OH , light blue) groups and two – groups. Here, we put forward the zipper model (which involves two – only) insteadin light of the computational results discussed in the “Results” section. HG crosslinking can also occur via hydrogen bonding, most prominently between methylated carboxyl groups – or even protonated carboxyl groups in very acidic conditions.
Specifically, we have investigated the molecular-level details of HG crosslinking via molecular simulations, quantifying the energetics of different cross-linking mechanisms. We have also systematically investigated the structure-function relationship between the functionalisation of HG chains and the kinetics of their aggregation. Our results suggests that the venerable “egg-box” model might not be adequate to describe the -mediated cross-linking in pectin. To investigate the impact of pectin cross-linking on the porosity of the cell wall, we employed an extended version of the fluorescence microscopy-based technique pioneered by Liu et al. . By visualizing changes in fluorescence from a fluorescently-labelled plasma membrane over time, we monitored the dynamics of a quenching molecule to penetrate the cell walls of wild type Arabidopsis thaliana plants and a mutant that is impaired in cell wall pectin cross-linking. In order to understand our experimental results, we used mathematical modelling based on fluid dynamics. Our results are consistent with the experimental data, provide a rationalisation of the time dependence of the fluorescence decay in terms of the morphology of the sample and lay the foundations for the future development of a predictive model – to quantify the porosity and potentially the tortuosity of the cell wall.

Results

Molecular simulations

The egg-box model of HG cross-linking is not supported. As the most abundant pectic domain in plant cell walls is , we focused on understanding the interactions between HG pectin chains. Our aim was to elucidate how pectin cross-links at the molecular level-which in turn might help us to understand the different degrees of porosity within the cell wall. To do this, we investigated the energetics of each type of HG-HG interaction: the main options in this regard are summarised in Fig. 1. Firstly, we consider the carboxyl groups of GalA, which in most cases we would expect to find in their de-protonated form (-COO- ), as their pKa is rather low (around 3.6). In this scenario, two – groups on two different chains can cross-link via a bridge.
Much of the existing literature suggests that these calcium bridges involve two hydroxyl (-OH) groups as well. This possibility, illustrated in Fig. 1, leads to the emergence of the so-called “egg-box” model for pectin cross-linking . This model is widely accepted amongst plant scientistshowever, it is difficult to resolve experimentally the exact position and-crucially-coordination of the ions within the cross-linking regions of HG chains .
Small-angle X-ray scattering (SAXS) has been previously employed to investigate the structural details of crosslinking in pectins and alginates . However, it is important to note that SAXS cannot provide direct evidence for the egg-box model. Notably, even recent experimental
work on arabidopsis cites previous modelling efforts, specifically the work of Braccini et al. , as evidence for the egg-box model. Interestingly, Braccini et al. do not advocate the original egg-box model but suggests a “shifted egg-box” configuration for galacturonate and guluronate chains.
In fact, recent work has called into question the validity of this model in favour of a “zipper” model , also illustrated in Fig. 1, which involves the two – groups only. This deceptively small difference between the egg-box and the zipper model does actually translate into a rather different configuration of the HG chain as a whole-as we will discuss in greater detail later.
In cases where the carboxyl group of GalA would be found in its protonated form, it is reasonable to expect some degree of hydrogen bonding between two – COOH groups belonging to different chains, without the need to involve a bridge . In fact, whilst-in presence of a sufficient amount of calcium- the formation of bridges is the main mechanism of HG cross-linking, in highly methyl- and /or acetylesterified pectin the cross-linking takes places thanks to hydrogen bonding, as illustrated in Fig. 1.
Thus, the resulting porosity of the pectin network within the cell wall is determined by a competition of different HG cross-linking mechanisms. In order to assess the stability of these different cross-linking options for HG chains, we performed metadynamics simulations. The details are discussed at length in the “Methods” section: in short, metadynamics is an enhanced sampling computational technique that leverages molecular dynamics (MD) simulations to reconstruct the free energy profiles relative to one or more “collettive variables”-in this case, specific degrees of freedom that describe the cross-linking process. To begin with, we have computed by means of unbiased simulations (i.e., without applying the metadynamics method so as not to tamper with the natural time evolution of the system) the running coordination number for in water (i.e. the average number of water molecules to be found within a given distance from the ion) so as to validate the reliability of our computational setup. The result, illustrated in Fig. 2a, indicates a rather extended first solvation shell (within 2.5 and ) that contains on average water molecules. This is consistent with previous results obtained via the CHARMM force field as well as first-principles MD simulations .
Via metadynamics simulations, we have investigated the free energy gain associated with the formation of a single bridge, which we report in Fig. 2b as a function of both the coordination number of with water and the coordination number of with any of the four oxygen atoms belonging to the two de-protonated carbonyl groups (see Fig. 2d). Our results indicate that is able to fully link with two groups only via the “zipper” mechanism discussed above, without the need to involve – OH groups via the “egg-box” mechanism. The most stable link involves all the four oxygen atoms belonging to the two de-protonated
Fig. 2 | Metadynamics simulations of HG cross-linking. a Running average coordination number of with water molecules in an unbiased MD simulation. The shaded region highlights the uncertainty associated with our estimate (min-max error); the inset illustrated a six-coordinated ion. b Free energy surface relative to the formation of a single – bridge, reported as a function of the coordination number of the ion with water molecules as well as the coordination number of the ion with any of the four oxygen atoms belonging to the two – groups . c Free energy profile relative to the interaction between a and a single group, reported as a function of the distance between the ion and any of the two oxygen atoms belonging to the group . Representative configuration of a bridge. The ion and the four oxygen
atoms belonging to the two groups are coloured in orange and green, respectively. We also show the three water molecules the ion is coordinated to in that specific configuration. e Free energy profile relative to the interaction between two methylated carboxyl groups, reported as a function of the distance between one of the two carbon atoms within the -CH 3 groups and one of the two oxygen atoms involved in the double bond; (a) typical configuration is depicted in (f), where the relevant carbon atoms are coloured in purple. Free energy profile relative to the interaction between two carboxyl groups, reported as a function of the distance between one of the two de-protonated oxygen atoms and one of the two hydrogen atoms belonging to the carboxyl groups; a typical configuration is depicted in (h), where the relevant hydrogen atoms are coloured in lime.
carbonyl groups, whilst the coordination number of the ion with water becomes three (as opposed to the value of 6.5 observed in solution). In fact, in the unbiased simulations of HG cross-linking that we will discuss later, we have never encountered a situation resembling the “egg-box” mechanism. The interaction between – and alone is very strong, as illustrated by the free energy profile reported in Fig. 2c. Here, we look at the free energy gain in bringing a closer to the – group-in particular, the collective variable of choice is the distance between and either of the two oxygen atoms within the group. The minimum of the free energy is characterised by , which is a value in excellent agreement with the previous estimate of the potential of mean force (a similar measure of energetics in this specific context) reported in the work of Assifaoui et al. .
In contrast, cross-linking of HG via hydrogen bonding seems to be leveraging much weaker interactions. This is quantified by the free energy profiles we have obtained relative to cross-linking via hydrogen bonding between either methyl-esterified GalA and – COOH groups, reported in Fig. 2e, g, respectively. The hydrogen bonding between – COOH groups is significantly stronger than that methylated carboxyl groups: versus . However, a single interaction is much stronger than either of these hydrogen bonds. It is worth noticing that the can only happen in rather acidic conditions to begin with, and it is only reasonable to expect the interaction between two acetylated GalA to be even weaker than that between two methylated carboxyl groups.
Thus, in light of these results we argue that the main cross-linking mechanism for low methoxyl pectin in physiological conditions for plants must rely on – bridges. In absence of sufficient and/or when dealing with high methoxyl pectin, hydrogen bonding interactions between methylated carboxyl groups might play a significant role,
but they are bound to result in less strongly linked pectin compared to the scenario. In turn, we argue that weaker HG-HG interactions are bound to result in more fluid pectin domains, thus ultimately leading to larger average pore size within the cell wall.
Impact of HG functionalisation on its cross-linking potential. To gain further insight into the kinetics of the HG cross-linking, we have also performed unbiased simulations of the aggregation of HG chains in water. In particular, we have considered HG chains containing either 8 or 40 GalA units, and we have systematically explored the impact of the functionalisation of the GalA units on the HG cross-linking. Our simulation boxes contained eight chains (either 8 – or 40 -unit long) featuring different combination of protonated -COOH groups ( or for 8- and 40-unit long chains, respectively, in Table 1), de-protonated – groups (D or d in Table 1) and methylated carboxyl groups ( or in Table 1). We summarise our results in Table 1, where we report the average number (computed over 40 ns -long MD simulations) of crosslinks (via calcium bridges or via either or hydrogen bonds) formed between the HG chains, the average lifetime of said cross-links and the average number of chains involved into the largest HG aggregate. We have verified the reproducibility of these results across 15, statistically independent MD simulations-as discussed in the Supplementary Information (SI; see Supplementary Table 1)
We find that D-only HG chains show the greatest extent of crosslinking, whilst P -only chains link much less effectively, and M-only chains even less so. This trend is not only evident in terms of the average number of cross-links, but the lifetime of the links is substantially longer for the bridges forming in D-only chains than for the hydrogen bonds formed in either P -only or M -only chains. These results are consistent with the energetics we have obtained by means of the
Table 1 | HG cross-linking as a function of functionalisation
System соон
PPPPPPPP 9.91(32) 1.43 6.03
DDDDDDDD 16.89 (32) 8.02 8
MMMMMMMM 6.06 (32) 1.65 3.94
DDPDPPDP 9.44 (16) 4.23 (16) 7.37 1.52 7.74
DDMDMMDM 8.66 (16) 1.71 (16) 7.88 1.53 7.35
DPDPDPDP 8.51 (16) 6.59 (16) 8.6 1.62 7.51
DMDMDMDM 9.09 (16) 2.21 (16) 5.53 1.97 6.98
DDPPDDPP 8.79 (16) 4.72 (16) 8.79 1.58 7.72
DDMMDDMM 8.46 (16) 3.52 (16) 8.43 1.57 7.54
PPDDDDPP 7.15 (16) 3.56 (16) 5.05 1.25 5.99
MMDDDDMM 8.16 (16) 1.71 (16) 7.11 1.33 4.81
DDPPPPDD 10.59 (16) 3.88 (16) 7.96 1.68 7.53
DDMMMMDD 9.49 (16) 2.98 (16) 6.35 2.09 7.92
DPPDDPPD 9.77 (16) 4.09 (16) 6.69 2.11 7.78
DMMDDMMD 9.77 (16) 1.75 (16) 8.62 1.37 7.38
PDDPPDDP 8.46 (16) 4.25 (16) 6.2 1.41 7.56
MDDMMDDM 7.95 (16) 5.56 (16) 8.96 1.89 7.63
DDDDPPPP 6.52 (16) 3.41 (16) 3.62 1.51 4.86
DDDDMMMM 7.80 (16) 2.40 (16) 5.41 1.66 3.85
dddd 58.05 (160) 10.99 7.95
mmmm 1.81 (160) 1.24 2.07
dddp 32.51 (120) 2.79 (40) 10.68 1.49 4.52
dddm 37.95 (120) 0.17 (40) 16.67 1.11 5.61
ddpd 36.12 (120) 0.37 (40) 12.41 1.04 6.07
ddmd 29.14 (120) 0.98 (40) 10.67 1.66 6.54
ddpp 32.07 (80) 8.15 (80) 11.93 1.94 5.14
ddmm 24.80 (80) 0.50 (80) 14.49 1.24 4.37
dppd 29.44 (80) 6.27 (80) 15.91 1.59 3.26
dmmd 28.85 (80) 2.23 (80) 11.00 1.90 3.44
pddp 21.24 (80) 8.44 (80) 11.06 1.69 3.17
mddm 16.83 (80) 0.92 (80) 9.80 1.63 2.63
dpdp 26.68 (80) 10.37 (80) 11.43 1.87 6.22
dmdm 31.44 (80) 3.88 (80) 14.63 1.61 5.39
dppp 15.29 (40) 14.96 (120) 10.28 1.82 3.45
dmmm 13.12 (40) 1.73 (120) 20.10 1.66 2.44
pdpp 12.24 (40) 7.92 (120) 14.53 1.45 1.93
mdmm 14.46 (40) 2.04 (120) 18.8 1.36 2.00
Our simulation boxes contained eight chains (either 8 – or 40-unit long) featuring different combinations of protonated -COOH groups (P or p for 8- and 40-unit long chains, respectively), de-protonated -COO- groups (Dord for 8 – and 40 -unit long chains, respectively) and methylated carboxyl groups ( or for 8 – and 40 -unit long chains, respectively). The results have been computed by averaging over 40 ns-long MD simulations. We report: the average number of cross-links via calcium bridges, ; the average number of cross-links via hydrogen bonds, see ; the average number of cross-links via-COOH hydrogen bonds, (see Fig. 1). We also report the average lifetime (in ns) of these cross-links as well as the average number of chains involved in the largest HG aggregate during the simulation, .
metadynamics simulations discussed above (and summarised in Fig. 2). In terms of the chains characterised by a mixture of either D and M or D and P units, it appears that 4 -unit “blocks” of different units (e.g., DDDDMMMM or DDDDPPPP) cross-link to a much lesser extent if compared to chains with the same composition where however the units are alternating along the chains (e.g., DMDMDMDM or DPDPDPDP). This trend might indicate that alternating D units might favour hydrogen bonding between M or P units as well. However, it seems that chains terminated by D units (i.e., DDPPPPDD or DDMMMMDD) cross-link to an even greater extent than chains alternating D and either M or P units.
These results are consistent across 8 -unit and 40 -unit chains: interestingly, the lifetime of the bridges is on average almost two times longer than that we have computed for the 8 -unit chains, whilst the lifetime relative to the hydrogen bonds remains largely unchanged. This is symptomatic of the collective effect realised by the formation of contiguous calcium bridges along the chains. In fact, our simulations allowed us to pinpoint quite clearly the occurrence of the “zipper” mechanism discussed above. In Fig. 3, we report a representative snapshot of the aggregation of eight D-only, 40 -unit HG chains. Despite the relatively short duration of the simulation, we can clearly see the formation of calcium
Fig. -mediated HG cross-linking. A representative snapshot of an unbiased MD simulation of (8) HG chains, 40 -unit each, taken after . Carbon, oxygen and hydrogen atoms within the three chains involved in the cross-linking are depicted in cyan, red and white, respectively. The ions involved in the crosslinking are depicted in orange. Note the emergence of the “zipper” mechanism (see text and Fig. 1).
bridges across multiple chains, without the need to involve any – OH group in the process (as advocated by the “egg-box” mechanism).
Overall, our simulations suggest that HG chains cross-link to each other according to this “zipper” mechanism that does not introduce significant torsion within the HG chain. In contrast, we note that for the -OH groups to be involved in the calcium bridges according to the “egg-box” model, the HG chains must distort to a significant extent. These considerations, whilst elucidating the molecular-level details of the cross-linking of HG chains, also suggest that the role of pH as well as the content of calcium within the cell wall might be key to determine the overall degree of cross-linking in pectin.
To better understand the influence of HG cross-linking on the porosity of the cell wall, we turn to the experimental measurements discussed in the next section. We remark at this stage that a large number of genes control pectin methyl esterification (and thus, have the potential to affect pectin cross-linking) and redundancy in their function means that genetic mutants are largely unaffected by loss of function of any single PME gene . In contrast, mutants that are affected globally in HG pectin methyl esterification exhibit extreme structural defects, including reduced growth and loss of cell adhesion, due to their effects on the HG-rich middle lamella . This makes it difficult to draw conclusions about cross-linking by comparing them with wild type plants.
To circumvent this issue whilst making appreciable alterations to pectin cross-linking, we have used a well-studied genetic mutant of Arabidopsis thaliana which exhibits qualitative differences in its RGII pectic domain but shows unaltered amounts of pectin and no defect in cell adhesion . This mutation decreases the formation of boron bridges between pectin chains. RG-II is covalently linked to HG pectin and in wild type plants, those boron bridges draw pectin chains closer together and facilitate further cross-linking via the mechanisms we have examined in our simulations . As the formation of boron bridges and their knock-on effect on the creation of other linkages between pectin chains, (e.g., HG- bridges) are intertwined and cannot be disentangled experimentally, we focus on understanding their collective impact on the porosity of the cell wall.

Experimental determination of the cell wall porosity in an arabidopsis cell-wall pectin cross-linking mutant

Recently, Liu et al. presented a fluorescence microscopy-based technique that measures porosity of plant and fungal cells by assessing the ability of relatively large fluorescence quenching molecules to traverse the cell wall. Fluorescent labelling of the plasma membrane allows the degree of penetration of the quencher to be calculated on the basis of its ability to quench the fluorescence. This technique is an elegant refinement of previously used methods to assess effective pore size. To test the effect of altered pectin crosslinking on cell wall porosity, we applied this technique, with some modifications, to wild type arabidopsis Col0 plants and a well-studied mutant, sensitive-to-freezing-8 (sfr8), which exhibits reduced pectin cross linking without exhibiting major developmental aberrations .
Central to our modification of the technique was the measurement of fluorescence over a series of time points, rather than through a single endpoint measurement, to reveal the dynamics of the process. Specifically, we labelled the plasma membrane of epidermal leaf peels with the fluorescent dye FM4-64 and introduced a quencher molecule, trypan blue. We then monitored the decline of the observed florescence of the samples over a 15 min period. As can be seen from the images in Fig. 4a, fluorescence levels were similar between the two genotypes before quencher application but declined more rapidly in the sfr 8 mutant than in the wild type. Note that this difference was significant at the moment of application (time value , and at time value and min value as well (see Fig. 4b). To investigate directly the pore size distributions in Col0 and sfr8 mutant plants we prepared samples from Col0 and sfr8 leaves for Scanning Electron Microscopy (SEM) measurements. Images (Fig. 4c) were analysed with FIJI to quantify both pore size (area), and the frequency of each pore size (Fig. 4d). Clear differences in pore size distribution between the two genotypes were observed. Specifically, the frequency distribution of the pore area reported in Fig. 4d indicates that sfr8 plants have, on average, a larger proportion of larger pores (pore area ) than the Col0 wild type and a smaller proportion of small pores (pore area ). These measurements allowed us to compute an estimate of the porosity of the samples as:
where is the average pore size, is the number of pores and is the total area of the sample. Whilst the leaf to leaf variation in pore sizes varied noticeably (hence it was not appropriate to compare mean pore sizes directly across the two genotypes), we gauge to be of the order of for both samples, with . The outcomes of these experimental measurements were then used to construct a mathematical model (see next section) to rationalise the connection between the time-dependency of the observed fluorescence and the average pore size of the cell wall of different samples.

Mathematical modelling of the porosity measurements

To gain further insight into the mechanistic aspects underpinning the experimental results reported in the previous section, we have constructed a minimalist mathematical model, based on fluid dynamics considerations. The variables and parameters used in this section are summarised in Table 2.
As illustrated in Fig. 5a, we consider a cell wall of area and thickness . The cell wall sits in between the plasma membrane, with area (same as the area of the cell wall) and thickness (where ), and a bath of fluid containing the quencher molecule. The concentration of the quencher is defined as , where the -coordinate runs perpendicular to the cell wall’s surface, whilst the concentration of the quencher in the bath, , is assumed to stay constant during the experiment. The concentration of the quencher in the membrane is assumed to be spatially homogeneous and as a function of time is defined as , with (i.e., the concentration of the quencher in the membrane at the start of the experiment is zero). The model assumes that the cell wall is structurally homogeneous (i.e., there is no dependence in on the properties of the wall), so that the concentration
Fig. 4 | Experimental measurements of the cell wall porosity in Col0 wild type and sfr8 arabidopsis leaves. a Change in the fluorescence of the FM4-64-stained plasma membrane (see text) for Col0 and sfr 8 cell wall samples. A time series of three representative fluorescence images reported for each point in time for each genotype are shown. The optical microscopy images were taken before quenching ( ), immediately after addition of trypan blue (F) and at 3 -min intervals thereafter. The scale bar (yellow) is . Change in the relative fluorescence as a function of time. The mean values of are measured every 3 min within a time interval, for both Col0 and sfr8. The value at the Y -axis intersection represents the unquenched sample ( ) and is equivalent to 1.0. Three separate Regions of Interest (ROI) were assessed per epidermal peel ( 39 and 42 ROI for Col0 and sfr8, respectively). The data were combined from three separate experiments. For
statistical comparisons, a two-way ANOVA with Bonferroni’s multiple comparisons test was employed. The values at time points 0,3 , and 6 min are 0.0141 , 0.003 , and 0.0221 , respectively. Other time points showed no significant difference. The “*” and “**” symbols indicate moderate ( value ) and strong ( value ) evidence, respectively, for a significant statistical difference between the two time series. c SEM images of Col0 and sfr8 cell walls. Three representative images, taken from 3 different leaves, are shown for both genotypes. The scale bar (yellow) is . d Pore size (area, ) frequency distributions for Col0 and sfr8. The pore size distributions from the ROI ( 9 and 11 ROI for Col0 and sfr8, respectively) of three individual plants per genotype was assessed. A frequency distribution with a bin width and range of 15 and , respectively, was employed.
gradient across it is:
The cell wall is assumed to be a porous medium, characterised by porosity (i.e., the ratio between the volume of empty space within the cell wall and the total volume of the cell wall) and tortuosity . The latter is a measure of the complexity of the pathway for the fluid (and thus, in our case, the quencher) to diffuse through the porous network of the cell wall. The quencher molecules are characterised by a diffusion coefficient in the bath. The values of reported in Table 2 have been estimated by means of MD simulations of a single quencher molecule in water at room temperature and ambient pressure (see “Methods” section). The diffusion coefficient of the quencher through the pores of the cell wall, however, is bound to be (much) lower than the diffusion coefficient in solution. As such, we assume that within a pore the quencher has diffusion coefficient , where (ref. 35). Here, we write down the “effective” diffusion coefficient of the quencher in the cell wall as :
We also assume that the quencher follows Fickian diffusion:
where is the diffusional flux (the amount of quencher diffusing in the direction per unit area per unit time). At time , the amount of quencher diffusing per unit time into the cell wall is . Thus, the total amount diffused by time is . This is diffusing into the membrane, and relating this to the concentration gives , and also .
Combining these equations, we get the ordinary differential equation:
with solution
This has characteristic timescale
Having found the concentration of the quencher in the membrane, we must now relate this to the fluorescence as measured in the experiments. The fluorescence at is denoted by , and the relative intensity defined as . This is related to the quencher concentration by the
Table 2|Variables and parameters utilised in the mathematical modelling of the porosity measurements
Name Symbol Units Reference Value
Time s
Concentration of quencher in membrane
Diffusional flux
Fluorescence intensity
Porosity
Tortuosity
Quencher rate coefficient in membrane ?
Decay timescale s ?
Quencher constant
Diffusion coefficient of quencher in bath D
Diffusion coefficient of quencher in pore
Wall thickness m
Membrane thickness m
Concentration of quencher in bath
Sample area known
Stern-Volmer equation :
which with (6) rearranges to:
This is the function used for the fitting of the experimental data reported in Fig. 5. We fit as a function of , which gives us values for and for each plant type.
In our case, from this fit we have obtained values for and . If we assume that the porosity (and possibly the tortuosity) are the only differences between the two samples, then:
To relate the tortuosity to the porosity, we use the simple Bruggeman expression . This assumes that the porous structure is homogeneous and isotropic, and formed of particles much smaller than the size of the sample. Using this, we obtain:
so that:
From the fit reported in Fig. 5 we get:
Fig. 5 | Mathematical modelling of the porosity measurements. a Schematics of the model. The relevant variables and parameters are reported in Table 2. b Fitting of the experimental data (see Fig. 4).
This result indicates that the cell wall of the sfr 8 mutant is more porous than the cell wall of the wild type. However, we stress that in order for this model to be able to provide quantitative prediction of the porosity and-crucially-the tortuosity of the cell wall of each sample, further experimental efforts are needed, aimed at, e.g., “calibrating” the model via measuring the time dependence of the fluorescence for samples of known porosity and/or tortuosity.
Nevertheless, despite the simplicity of our model, our results indicate that: (i.) the experimental results can be explained without the need of introducing any “blocking” mechanism for the quencher to get stuck in and/ or occlude the pores of the cell wall ; (ii.) measuring the decay in the fluorescence of the sample as early as possible during the experiments is justified by the fact that by doing so one maximises the impact of the different porosity of the samples; (iii.) the different porosity and/or tortuosity of the samples alone is sufficient to explain the difference in relative intensity observed experimentally. Future work from the modelling standpoint might explore more sophisticated models for the tortuosity, as this parameter provides information about the morphology of the pore network.

Discussion

The plant cell wall is a highly complex structure governed by numerous interactions between carbohydrate polymers and structural proteins including extensins . These interactions, particularly the formation of cross-links between pectin chains, determine not only the physical strength of the structure but its porosity, which impacts upon the ease of entry of a range of small, biologically important macromolecules, including enzymes, cell wall structural proteins, intercellular signalling molecules and carbohydrates .
To gain a better understanding of what shapes porosity in plant cell walls, we began with the pectin building blocks and examined the possible molecular interactions between pectin chains that are capable of crosslinking pectin and making it less porous. Our MD simulations added evidence to an emerging case to be made for a zipper model rather than the popular egg box model favoured by many plant biologists. Interestingly, our MD simulations also highlighted the importance of pH on the wall and
cross-linking. These data are consistent with experimental observations in vitro that pectin gelling and firmness is dependent upon both and . In fact, the acidity of the wall has long been appreciated as determining growth and extensibility of the wall for a variety of reasons, which include the pH dependence of cross-linking of .
Having investigated the molecular basis for the different ways pectin chains can aggregate and interact with one another, we gathered experimental data from a biological system in which pectin cross-linking is altered. We used time-dependent fluorescence quenching measurements to assess how altered cross-linking affected the porosity of cell wall. We confirmed the validity of these measurements with direct observations via SEM and we modelled the fluorescence quenching data as well. Whilst arabidopsis mutants in HG pectin methyltransferases are available and would be predicted to exhibit altered crosslinking, these suffer from numerous defects including stunted growth and a loss of cell adhesion . Thus, to obtain the experimental data for our porosity modelling, we used a more amenable arabidopsis mutant, which shows no cell adhesion defects and minimal effects on growth . Sensitive-to-freezing-8 (sfr8) harbours a mutation in the well-studied murl gene, known to influence RGII pectin cross-linking . Due to a modification of the sugar side chain A, RGII pectin cross-linking is reduced in sfr8/mur1 mutants. In wild type arabidopsis cell walls, RGII pectic domains can cross-link via borate diester linkages. RGII is covalently linked to HG pectin so in turn, these linkages bring the HG chains closer to each other, thus resulting in an enhanced probability for HG chains to further cross-link to each other. This is borne out of experimental observations that demonstrate the dependency of HG- cross-linking on RGII boron bridges. Pectin was more vulnerable to EDTA extraction in murl mutants, which exhibit relocation of non-methylesterified . Furthermore, boron-bridged RG-II and calcium are both required to maintain the pectin network of the arabidopsis seed mucilage ultrastructure and affect pectin properties including viscosity. The marked differences in porosity observed in cell walls with and without boron cross-linking of this relatively minor pectic domain would support the idea that reducing RGII boron bridges has a downstream and greater widespread effect on HG crosslinks also.
To investigate this hypothesis, we used a modified version of the fluorescence quenching method developed by Liu et al. , to probe the porosity of wild type arabidopsis and sfr8 mutant cell walls. As expected, our results showed that the quenching molecule traversed the cell wall to quench the fluorescence in the membrane more quickly in sfr8 than in Col 0 wild type, consistent with reduced cross-linking. To validate our results, we used SEM to image the cell wall matrices of leave and assess the amount of visible pore space are in each. Consistent with our fluorescence results, this confirmed that in the mutant, the distribution of size areas is altered: sfr8 exhibits fewer smaller pore areas and a greater number of larger pore areas.
In order to rationalise the experimental fluorescence measurements, and particularly to elucidate the physical processes underpinning the results, we have built a straightforward mathematical model leveraging fluid dynamics concepts. Our results suggests that the flux of the quencher molecules through the cell wall does not occlude the pores within the pectin network; although in future, modifications to this model which include a “blocking mechanism” could be considered when dealing with larger quencher molecules.
We also provide a justification for the up-to-now empirical choice of measuring the fluorescence decay within the sample as soon as possible after having introduced the quencher. This is because the early time regimes maximise the difference in terms of quenching efficiency. Finally, we demonstrate that the different responses in terms of fluorescence elicited by different samples can be explained in terms of the porosity of the samples alone.
The ability to predict the interactions between pectin chains in different contexts and forecast changes in porosity has many applications in agritechnology. Pectin structure determines the level of success attacking fungal and bacterial pathogens may have and mechanisms have even been identified whereby the invading fungus targets Ca-pectate domains to
render the wall more porous and facilitate access to the cell’s nutrients . Growing evidence suggests that remodelling of cell wall and particularly pectin structure is involved in the defence against abiotic stresses and there are reports of wall porosity affecting susceptibility to ion toxicity and dehydration . Altering the porosity of cell wall has also been shown to increase the efficiency of uptake, increasing photosynthesis as a result . In addition, differences in cell wall porosity can be exploited in optimization of nanoparticle uptake , wood quality and strength and saccharification of plant and algal material to produce biofuels . There are clearly opportunities for crop protection here.
In summary, this study shed light on the intricate dynamics of pectin cross-linking and its consequential effects on plant cell wall porosity, offering significant insights across multiple disciplines. By elucidating the molecular interactions within pectin chains and their impact on cell wall structure, we have advanced our understanding of plant biology, particularly in relation to plant growth, disease resistance, and stress responses. Our findings highlight the potential for manipulating pectin structure to enhance crop resilience against pathogens and abiotic stress or improve saccharification potential, pointing to novel strategies for crop protection and improvement. This cross-disciplinary research not only deepens our fundamental understanding of plant cell wall chemistry but also opens new avenues for agricultural innovation and sustainability.

Methods

Growth of plants

Arabidopsis plants were grown under 12 h light: 12 h dark at , in a walk-in growth room as described previously . sensitive-to-freezing-8 (sfr8) and Columbia-0 (Col0) wild type plants were compared. Plants were grown to the rosette leaf stage (approximately 5 weeks post germination). Leaves were removed for porosity measurement or analysis by scanning electron microscopy.

Porosity measurements

Fluorescence measurements. The porosity of the cell wall in leaf epidermal peels was assessed through a fluorescence quenching assay, following the method originally described by Liu et al. . We used the detailed protocol outlined for maize leaves in the work of Liu et al. with the following modifications. Abaxial epidermal peels were obtained from 5-week-old plants using the Perforated-tape Epidermal Detachment (PED) technique described by Ibata et al. . The peels were incubated for 5 mins in a M FM4-64 (Thermo Fisher Scientific) solution. Following incubation, the epidermal peel was rinsed with water and then mounted on a microscope glass slide. Before mounting, Scotch Tape was carefully applied along the long edges of the glass slide, creating a chamber in the centre where the sample was placed and covered with a cover slip. An initial image ( ) was captured prior to the addition of trypan blue (Sigma-Aldrich) to quench the fluorescence of the sample. Subsequently, a series of images ( ) were acquired at intervals over a span of 15 mins. Imaging was carried out using a Zeiss 800 laser scanning confocal microscope equipped with a 1.4 NA 63x oil immersion objective, using a 488 nm excitation laser and fluorescence detection at . The ZEN Blue imaging software provided by the microscope manufacturers (Zeiss) was utilized for image acquisition, while analysis was done using ImageJ software . For image analysis, the mean intensity of a defined region of interest (ROI), specifically a rectangle, was quantified. Three separate ROIs were assessed for each sample. Intensity values were exported, and the relative porosity calculated as .
Electron microscopy. Chemicals were obtained from Sigma, Poole, UK or Fisher Scientific (unless otherwise stated). A single leaf from a 5 -week old arabidopsis plant was placed in PEM buffer ( 25 mM PIPES [1,4-piperazinediethanesulfonic acid], ) at room temperature and cut to a width of in the buffer. The tissue sample
was transferred to PEM buffer with formaldehyde (TAAB, Aldermaston, UK) and placed on ice for . Samples were then transferred to Karnovsky’s fix ( glutaraldehyde (TAAB), 2% formaldehyde in PEM buffer) and incubated overnight at . The following day, samples were washed 3-5 times for 5-10 min in PEM buffer, before incubating in 0.5% sodium hypochlorite in PEM buffer for 20 mins at room temperature. After five 10 -min rinses in PEM buffer, samples were transferred to osmium tetroxide (Agar Scientific Ltd., Stansted, UK) in PEM buffer, for 60-120 min at . Samples were dehydrated using an ethanol series with at in each of the following: ethanol. Dehydration was completed using 3 successive washes in ethanol for 10 min each at room temperature. After critical point drying (CPD) in a Leica Critical Point Dryer EM CPD030, the sample was attached to a piece of carbon double-sided adhesive tape (aluminium base; NISSHIN-EM CO., LTD) attached to a silicon mount (Agar Scientific). The sample was then surrounded by silver paint (Micro to Nano, Haarlem, Netherlands) to improve the conductivity between the sample and the silicon wafer and sputter coated with approximately 2 nm of platinum at a 45 degree angle, using a 328UHR coating unit (Cressington Scientific instruments, Watford, UK). Pore size was determined from leaves obtained from three individual plants for Col0 and four individual plants for sfr8, utilizing the methodology outlined by Hojat et al. . In brief, SEM images were processed using ImageJ software , where a region of was selected from each image for analysis. Initially, thresholding was applied to segment the images into pore and non-pore regions, followed by the application of morphological filters, erode and dilate. The size and quantity of pores were extracted from each image and converted to nanometres. Pore areas falling within the range of were utilized to generate a frequency distribution graph, while those outside this range were excluded from the analysis.

Molecular dynamics simulations

Molecular dynamics simulations were performed in GROMACS 5.1.3 , using the all-atom CHARMM36 forcefield together with either the TIP4P water model or the TIP4P/Ice water model . Initial simulations were carried out using TIP4P/Ice, as this model has been shown to be particularly effective for simulations of biomolecules in supercooled water . Due to the dynamics of these systems being extremely slow in TIP4P/Ice, and therefore difficult to explore on the simulation timescale, we have chosen to switch to the TIP4P water model, which at room temperature is characterised by a much faster mobility. This choice allowed us to investigate the aggregation of the HG chains without the need to resort to enhanced sampling simulations – too computationally expensive to deal with the systematic investigation we have conducted here. Additional details with respect to this computational strategy can be found in the SI, Supplementary Fig. S1.
Simulations have been run on a range of systems with different HG chain compositions (see Table 1). Each system consisted of eight identical chains with eight residues each, which were either protonated (P), deprotonated (D), or methylesterified (M). Deprotonated residues carry a negative ( ) charge, so an appropriate number of calcium ( ) ions were added to systems containing chains with deprotonated residues to balance the overall charge. The protocol for these simulations was as follows. The HG chains were placed randomly into a cubic simulation box of initial side 10 nm . The system was then solvated in water, and if necessary, some water molecules were replaced with calcium ions to balance the charge. An energy minimisation was then carried out using a steepest descent algorithm , followed by a 60 ns run, with a timestep of 2 fs , at room temperature and ambient pressure in the NPT ensemble. The Bussi-Donadio-Parrinello thermostat and Berendsen barostat were used, with coupling constants of 0.5 and 4 ps , respectively. Periodic boundary conditions were applied in three dimensions. Additional computational details with respect to these MD simulations can be found in the SI, Supplementary Table 2.
To investigate in detail the energetics of the different interactions illustrated in Fig. 1, we have resorted to well-tempered metadynamics
simulations using PLUMED 2.4.2 . Well-tempered metadynamics is an enhanced sampling technique that allows the free energy surface of system to be explored relative to a chosen set of degrees of freedom, or “collective variables” (CVs). These simulations have been used to measure the relative strengths of the two HG linking methods, namely calcium bridges and hydrogen bonds. The simulations contained two identical HG chains of eight residues, with seven protonated residues and either one deprotonated residue or one methylesterified residue. For the calcium bridge, the distance between the calcium ion and the deprotonated site was chosen as the CV to bias: the width, height and deposition stride of the Gaussian potentials were set (after extensive testing and validation) to , and 500 steps, respectively. The bias factor was set to . For the hydrogen bond, the distance between the methyl carbon on one residue and the carbonyl oxygen on the other residue was chosen as the CV to bias: the width, height and deposition stride of the Gaussian potentials were set (after extensive testing and validation) to , and 500 steps, respectively. The bias factor was set to . Additional computational details with respect to these metadynamics simulations can be found in the SI, Supplementary Table 3.
The deposition rate of the bias was set to 500 MD steps (equivalent to 1 ps ) as the structural relaxation time of TIP4P water at room temperature and pressure is of the order of . The width of the Gaussian potentials was initially set to . This value turned out to be too small, in that the resulting free energy surfaces struggled to achieve convergence due to the too-fine resolution. Thus, we have increased to 0.04 and 0.05 nm for investigating the calcium bridge and the hydrogen bond interactions, respectively. Values of resulted in free energy surfaces lacking sufficient detail/resolution. was initially set to half the value of the thermal energy ( ) at room temperature, i.e., . This is a common rule of thumb in terms of striking a compromise between sampling efficiency and accuracy. We have explored values of between 1.00 and , in conjunction with bias factors (which value also influences said efficiency/ accuracy ratio) ranging from 50 to 100 . Any combination of the values of and within the above-mentioned ranges led us to obtain free energy surfaces within the uncertainty computed via the reweighing technique of Tiwary and Parrinello (and reported in Fig. 2 as the shaded blue regions) within an acceptable simulation time ( ).
Whilst the concept of time in metadynamics simulations does not have a direct physical meaning (as the time evolution of the system is heavily influenced by the artificial bias introduced for the purposes of exploring the free energy surface of interest), such long simulation times were necessary to ensure the proper convergence of the resulting free energy surface. To this end, we have: (i.) monitored the changes of the free energy surface as a function of the simulation time; (ii.) explored the impact of varying the height and width of the bias potential; (iii.) utilised the reweighing technique of Tiwary and Parrinello to provide a quantitative estimate of the uncertainty associated with the free energy surface.
Statistics and Reproducibility. We have included a discussion of the reproducibility of our results with respect to every aspect of our methodology (see the Methods section). Perhaps the most delicate aspect of our work concerns the reproducibility of our results in terms of the molecular dynamics simulations of HG aggregates (see Table 1). To this end, we have included direct evidence of the robustness of these results in the Supplementary Information. With respect to the fluorescence measurements reported in Fig. 4b, three separate regions of interest (ROI) were assessed per epidermal peel (for a total of 39 and 42 ROI for Col 0 and sfr8, respectively). The data were combined from three separate experiments. For statistical comparisons, a two-way ANOVA with Bonferroni’s multiple comparisons test was employed. Significance was set at . In terms of the SEM measurements, the pore size distributions from the ROI ( 9 and 11 ROI for Col0 and sfr8, respectively) of three individual plants per genotype was assessed. A frequency distribution with a bin width and range of 15 and , respectively, was employed. Analyses were performed using GraphPad Prism version 10.1.0 for Windows. In terms of the results of the metadynamics
simulations reported in Fig. 2, we have utilized the reweighing technique of Tiwary and Parrinello to assess the uncertainty associated with the resulting free energy profiles.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All of the data associated with this study are available via a publicly accessible GitHub repository (PEC_LINK, https://github.com/gcsosso/PEC_LINK. git). We have also linked a release of said repository to a Zenodo repository (also publicly accessible), for which we have obtained a permanent DOI . Any remaining information can be obtained from the corresponding author upon reasonable request.

Code availability

The details of the codes utilized in this work can be found in the Methods section. No in-house or proprietary code has been utilized. With respect to the computational aspects of the work, we remark that both the GROMACS and the PLUMED packages are open source and publicly available.
Received: 8 April 2024; Accepted: 8 January 2025;
Published online: 17 January 2025

References

  1. Cosgrove, D. J. Growth of the plant cell wall. Nat. Rev. Mol. Cell Biol. 6, 850-861 (2005).
  2. Delmer, D., Dixon, R.A., Keegstra, K., Mohnen, D. The plant cell walldynamic, strong, and adaptable-is a natural shapeshifter. Plant Cell, 325. https://doi.org/10.1093/plcell/koad325 (2024).
  3. Caffall, K. H. & Mohnen, D. The structure, function, and biosynthesis of plant cell wall pectic polysaccharides. Carbohydr. Res. 344, 1879-1900 (2009).
  4. Baron-Epel, O., Gharyal, P. K. & Schindler, M. Pectins as mediators of wall porosity in soybean cells. Planta 175, 389-395 (1988).
  5. Fleischer, A., O’Neill, M. A. & Ehwald, R. The pore size of nongraminaceous plant cell walls is rapidly decreased by borate ester cross-linking of the pectic polysaccharide rhamnogalacturonan II1. Plant Physiol. 121, 829-838 (1999).
  6. Bidhendi, A. J. & Geitmann, A. Relating the mechanics of the primary plant cell wall to morphogenesis. J. Exp. Bot. 67, 449-461 (2016).
  7. Carpita, N., Sabularse, D., Montezinos, D. & Delmer, D. P. Determination of the pore size of cell walls of living plant cells. Science 205, 1144-1147 (1979).
  8. Pérez-Donoso, A. G., Sun, Q., Roper, M. C., Greve, L. C., Kirkpatrick, B. & Labavitch, J. M. Cell wall-degrading enzymes enlarge the pore size of intervessel pit membranes in healthy and Xylella fastidiosainfected grapevines. Plant Physiol. 152, 1748-1759 (2010).
  9. Savary, S., Willocquet, L., Pethybridge, S. J., Esker, P., McRoberts, N. & Nelson, A. The global burden of pathogens and pests on major food crops. Nat. Ecol. Evol. 3, 430-439 (2019).
  10. Adani, F., Papa, G., Schievano, A., Cardinale, G., D’Imporzano, G. & Tambone, F. Nanoscale structure of the cell wall protecting cellulose from enzyme attack. Environ. Sci. Technol. 45, 1107-1113 (2011).
  11. Wang, Y., Xu, X., Xue, H., Zhang, D. & Li, G. Physical-chemical properties of cell wall interface significantly correlated to the complex recalcitrance of corn straw. Biotechnol. Biofuels 14, 196 (2021).
  12. Grant, G. T., Morris, E. R., Rees, D. A., Smith, P. J. C. & Thom, D. Biological interactions between polysaccharides and divalent cations: The egg-box model. FEBS Lett. 32, 195-198 (1973).
  13. Zdunek, A., Pieczywek, P. M. & Cybulska, J. The primary, secondary, and structures of higher levels of pectin polysaccharides. Compr. Rev. Food Sci. Food Saf. 20, 1101-1117 (2021).
  14. Assifaoui, A. et al. Structural behaviour differences in low methoxy pectin solutions in the presence of divalent cations (Ca2+ and Zn2+): a process driven by the binding mechanism of the cation with the galacturonate unit. Soft Matter 11, 551-560 (2014).
  15. Poset, A. M. D., Zitolo, A., Cousin, F., Assifaoui, A. & Lerbret, A. Evidence for an egg-box-like structure in iron(II)-polygalacturonate hydrogels: a combined EXAFS and molecular dynamics simulation study. Phys. Chem. Chem. Phys. 22, 2963-2977 (2020).
  16. Pelloux, J., Rustérucci, C. & Mellerowicz, E. J. New insights into pectin methylesterase structure and function. Trends Plant Sci. 12, 267-277 (2007).
  17. Harholt, J., Suttangkakul, A. & Vibe Scheller, H. Biosynthesis of pectin. Plant Physiol. 153, 384-395 (2010).
  18. Wormit, A. & Usadel, B. The multifaceted role of pectin methylesterase inhibitors (PMEls). Int J. Mol. Sci. 19, 2878 (2018).
  19. O’Neill, M. A., Ishii, T., Albersheim, P. & Darvill, A. G. RHAMNOGALACTURONAN II: structure and function of a borate cross-linked cell wall pectic polysaccharide.Annu. Rev. Plant Biol. 55, 109-139 (2004).
  20. Rondeau-Mouro, C., Defer, D., Leboeuf, E. & Lahaye, M. Assessment of cell wall porosity in Arabidopsis thaliana by NMR spectroscopy. Int. J. Biol. Macromol. 42, 83-92 (2008).
  21. Liu, X. et al. Novel tool to quantify cell wall porosity relates wall structure to cell growth and drug uptake. J. Cell Biol. 218, 1408-1421 (2019).
  22. Fu, J., Thomas, H. R. & Li, C. Tortuosity of porous media: Image analysis and physical simulation. Earth Sci. Rev. 212, 103439 (2021).
  23. Du, J., Anderson, C. T. & Xiao, C. Dynamics of pectic homogalacturonan in cellular morphogenesis and adhesion, wall integrity sensing and plant development. Nat. Plants 8, 332-340 (2022).
  24. Plazinski, W. Molecular basis of calcium binding by polyguluronate chains. Revising the egg-box model. J. Comput Chem. 32, 2988-2995 (2011).
  25. Temple, H. et al. Golgi-localized putative S-adenosyl methionine transporters required for plant cell wall polysaccharide methylation. Nat. Plants 8, 656-669 (2022).
  26. Draget, K. I., Stokke, B. T., Yuguchi, Y., Urakawa, H. & Kajiwara, K. Small-angle X-ray scattering and rheological characterization of alginate gels. 3. alginic acid gels. Biomacromolecules 4, 1661-1668 (2003).
  27. Ventura, I., Jammal, J. & Bianco-Peled, H. Insights into the nanostructure of low-methoxyl pectin-calcium gels. Carbohydr. Polym. 97, 650-658 (2013).
  28. Shi, D.-C. et al. Boron-bridged RG-II and calcium are required to maintain the pectin network of the Arabidopsis seed mucilage ultrastructure. Plant Mol. Biol. 94, 267-280 (2017).
  29. Braccini, I. & Pérez, S. Molecular basis of Ca2+-induced gelation in alginates and pectins: the egg-box model revisited. Biomacromolecules 2, 1089-1096 (2001).
  30. Gilsenan, P. M., Richardson, R. K. & Morris, E. R. Thermally reversible acid-induced gelation of low-methoxy pectin. Carbohydr. Polym. 41, 339-349 (2000).
  31. Yoo, J., Wilson, J. & Aksimentiev, A. Improved model of hydrated calcium ion for molecular dynamics simulations using classical biomolecular force fields. Biopolymers 105, 752-763 (2016).
  32. Panter, P. E. et al. Cell wall fucosylation in Arabidopsis influences control of leaf water loss and alters stomatal development and mechanical properties. J. Exp. Bot. 74, 2680 (2023).
  33. Ishii, T. & Matsunaga, T. Pectic polysaccharide rhamnogalacturonan II is covalently linked to homogalacturonan. Phytochemistry 57, 969-974 (2001).
  34. Panter, P. E. et al. MUR1-mediated cell-wall fucosylation is required for freezing tolerance in Arabidopsis thaliana. N. Phytol. 224, 1518-1531 (2019).
  35. Bruna, M. & Chapman, S. J. Diffusion of finite-size particles in confined geometries. Bull. Math. Biol. 76, 947-982 (2014).
  36. Tartakovsky, D. M. & Dentz, M. Diffusion in porous media: phenomena and mechanisms. Transp. Porous Media 130, 105-127 (2019).
  37. Lakowicz, J.R. (ed). Principles of Fluorescence Spectroscopy. https:// doi.org/10.1007/978-0-387-46312-4 (Springer, 2006).
  38. Tjaden, B., Cooper, S. J., Brett, D. J., Kramer, D. & Shearing, P. R. On the origin and application of the Bruggeman correlation for analysing transport phenomena in electrochemical systems. Curr. Opin. Chem. Eng. 12, 44-51 (2016).
  39. Herterich, J. G., Griffiths, I. M. & Vella, D. Reproducing the pressure-time signature of membrane filtration: The interplay between fouling, caking, and elasticity. J. Membr. Sci. 577, 235-248 (2019).
  40. Castilleux, R., Plancot, B., Vicré, M., Nguema-Ona, E. & Driouich, A. Extensin, an underestimated key component of cell wall defence? Ann. Bot. 127, 709-713 (2021).
  41. Read, S.M., Bacic, A. Cell wall porosity and its determination. in Plant Cell Wall Analysis (eds Linskens, H.F., & Jackson, J.F.) 63-80. https:// doi.org/10.1007/978-3-642-60989-3_4 (Springer, 1996).
  42. Löfgren, C., Guillotin, S., Evenbratt, H., Schols, H. & Hermansson, A.M. Effects of calcium, pH, and blockiness on kinetic rheological behavior and microstructure of HM pectin gels. Biomacromolecules 6, 646-652 (2005).
  43. Yang, X. et al. Low methoxyl pectin gelation under alkaline conditions and its rheological properties: Using NaOH as a pH regulator. Food Hydrocoll. 79, 560-571 (2018).
  44. Phyo, P., Gu, Y. & Hong, M. Impact of acidic pH on plant cell wall polysaccharide structure and dynamics: insights into the mechanism of acid growth in plants from solid-state NMR. Cellulose 26, 291-304 (2019).
  45. Mouille, G. et al. Homogalacturonan synthesis in Arabidopsis thaliana requires a golgi-localized protein with a putative methyltransferase domain. Plant J. 50, 605-614 (2007).
  46. Kim, S.-J., Held, M. A., Zemelis, S., Wilkerson, C. & Brandizzi, F. CGR2 and CGR3 have critical overlapping roles in pectin methylesterification and plant growth in Arabidopsis thaliana. Plant J. 82, 208-220 (2015).
  47. Du, J. et al. Mutations in the pectin methyltransferase QUASIMODO2 influence cellulose biosynthesis and wall integrity in Arabidopsis. Plant Cell 32, 3576-3597 (2020).
  48. O’Neill, M. A., Eberhard, S., Albersheim, P. & Darvill, A. G. Requirement of borate cross-linking of cell wall rhamnogalacturonan II for Arabidopsis growth. Science 294, 846-849 (2001).
  49. Bellincampi, D., Cervone, F., Lionetti, V. Plant cell wall dynamics and wall-related susceptibility in plant-pathogen interactions. Front. Plant Sci. 5, 228 (2014).
  50. Forand, A. D. et al. With a little help from my cell wall: structural modifications in pectin may play a role to overcome both dehydration stress and fungal pathogens. Plants 11, 385 (2022).
  51. Vogel, J. P., Raab, T. K., Schiff, C. & Somerville, S. C. PMR6, a pectate lyase-like gene required for powdery mildew susceptibility in Arabidopsis. Plant Cell 14, 2095-2106 (2002).
  52. Tenhaken, R. Cell wall remodeling under abiotic stress. Front. Plant Sci. 5, 771 (2015).
  53. Panter, P.E., Panter, J.R., Knight, H. Impact of cell-wall structure and composition on plant freezing tolerance. In Annual Plant Reviews Online 607-642. https://doi.org/10.1002/9781119312994.apr0746 (John Wiley & Sons, 2020).
  54. Wu, H.-C., Bulgakov, V. P. & Jinn, T.-L. Pectin methylesterases: cell wall remodeling proteins are required for plant response to heat stress. Front. Plant Sci. 9, 1612 (2018).
  55. Yang, Z.-B., Eticha, D., Rao, I. M. & Horst, W. J. Alteration of cell-wall porosity is involved in osmotic stress-induced enhancement of
    aluminium resistance in common bean (Phaseolus vulgaris L.). J. Exp. Bot. 61, 3245 (2010).
  56. Dang, Z. et al. The Fragile culm19 (FC19) mutation largely improves plant lodging resistance, biomass saccharification, and cadmium resistance by remodeling cell walls in rice. J. Hazard. Mater. 458, 132020 (2023).
  57. Salesse-Smith, C. E. et al. Greater mesophyll conductance and leaf photosynthesis in the field through modified cell wall porosity and thickness via AtCGR3 expression in tobacco. Plant Biotechnol. J. 22, 2504-2517 (2024).
  58. Wu, H. & Li, Z. Nano-enabled agriculture: How do nanoparticles cross barriers in plants? Plant Commun. 3, 100346 (2022).
  59. Li, X. & Zhao, Z. Time domain-NMR studies of average pore size of wood cell walls during drying and moisture adsorption. Wood Sci. Technol. 54, 1241-1251 (2020).
  60. Loqué, D., Scheller, H. V. & Pauly, M. Engineering of plant cell walls for enhanced biofuel production. Curr. Opin. Plant Biol. 25, 151-161 (2015).
  61. Wang, Y., Xu, X., Xue, H., Zhang, D., Li, G. Recalcitrance of corn straw significantly correlated to the heterogeneity in physicochemical properties of cell wall. Res. Sq. https://doi.org/10.21203/rs.3.rs-605820/v1 (2021).
  62. Liu, X., Pomorski, T. G. & Liesche, J. Non-invasive quantification of cell wall porosity by fluorescence quenching microscopy. Bio. Protoc. 9, 3344 (2019).
  63. Ibata, H., Nagatani, A. & Mochizuki, N. Perforated-tape epidermal detachment (PED): a simple and rapid method for isolating epidermal peels from specific areas of Arabidopsis leaves. Plant Biotechnol. 30, 497-502 (2013).
  64. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676-682 (2012).
  65. Sugimoto, K., Williamson, R. E. & Wasteneys, G. O. New techniques enable comparative analysis of microtubule orientation, wall texture, and growth rate in intact roots of Arabidopsis. Plant Physiol. 124, 1493-1506 (2000).
  66. Hojat, N., Gentile, P., Ferreira, A. M. & Šiller, L. Automatic pore size measurements from scanning electron microscopy images of porous scaffolds. J. Porous Mater. 30, 93-101 (2023).
  67. Abraham, M. J. et al. Gromacs: high performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1-2, 19-25 (2015).
  68. Guvench, O. et al. CHARMM additive all-atom force field for carbohydrate derivatives and its utility in polysaccharide and carbohydrate- protein modeling. J. Chem. Theory Comput. 7, 3162-3180 (2011).
  69. Lawrence, C. P. & Skinner, J. L. Flexible TIP4P model for molecular dynamics simulation of liquid water. Chem. Phys. Lett. 372, 842-847 (2003).
  70. Abascal, J.L.F., Sanz, E., Fernández, R.G., Vega, C. A potential model for the study of ices and amorphous water: TIP4P/Ice. J. Chem. Phys. 122, 234511 (2005).
  71. Lee, H. Interactions of antifreeze proteins in TIP4P / Ice water and their dependence on force fields. PLoS ONE 07, 1-17 (2018).
  72. Mochizuki, K. & Molinero, V. Antifreeze glycoproteins bind reversibly to ice via hydrophobic groups. J. Am. Chem. Soc. 140, 4803-4811 (2018).
  73. Sosso, C. G. et al. Unravelling the origins of ice nucleation on organic crystals. Chem. Sci. 9, 8077-8088 (2018).
  74. Mayne, D. Q. A stochastic steepest-descent algorithm. J. Optim. Theory Appl. 59, 307-323 (1988).
  75. Bussi, G., Donadio, D. & Parrinello, M. Canonical sampling through velocity rescaling. J. Chem. Phys. 126, 1-7 (2007).
  76. Berendsen, H. J. C., Postma, J. P. M., van Gunsteren, W. F., DiNola, A. & Haak, J. R. Molecular dynamics with coupling to an external bath. J. Chem. Phys. 81, 3684-3690 (1984).
  77. Laio, A. & Parrinello, M. Escpaing free-energy minima. PNAS 99, 12562-12566 (2002).
  78. Barducci, A. & Bussi, G. Well-tempered metadynamics: a smoothly converging and tunable free- energy method. Phys. Rev. Lett. 100, 1-4 (2008).
  79. Tribello, G. A., Bonomi, M., Branduardi, D., Camilloni, C. & Bussi, G. PLUMED 2: new feathers for an old bird. Comput. Phys. Commun. 185, 604-613 (2014).
  80. Gallo, P. & Rovere, M. Relation between the two-body entropy and the relaxation time in supercooled water. Phys. Rev. E 91, 012107 (2015).
  81. Tiwary, P. & Parrinello, M. A time-independent free energy estimator for metadynamics. J. Phys. Chem. B 119, 736-742 (2015).
  82. McHugh, M. L. Multiple comparison analysis testing in ANOVA. Biochem. Med. 21, 203-209 (2011).
  83. Sosso, G.C. PEC_LINK. Accessed: 2025-01-08 (2024). https:// zenodo.org/records/14366068

Acknowledgements

This work has been supported by the Biotechnology and Biological Sciences Research Council (BBSRC) grant “Understanding ice formation in plants: finding new routes to freezing tolerance (Plantlce)” (BB/N015559/1). N.R. was funded by a BBSRC Doctoral Training Partnership studentship (BB/ M011186/1). P.L.J. is funded via the EPSRC CDT in Modelling of Heterogeneous Systems (HetSys, grant number EP/S022848/1). J.E.S acknowledges support from the EPSRC under grants EP/W031426/1, EP/ S029966/1 and EP/P031684/1. I.J.P., F.B. and G.C.S. gratefully acknowledge the usage of the ARCHER2 UK National Supercomputing Service (https://www.archer2.ac.uk), which we have accessed via the HecBioSim consortium (project e676), funded by the EPSRC (Grant No. EP/ X035603/1). I.J.P. and G.C.S. also gratefully acknowledge the use of SULIS, which was funded by the EPSRC (EP/T022108/1), via the HPC Midlands+ Consortium. I.P., F.B. and G.C.S. would also like to acknowledge the highperformance computing facilities provided by the Scientific Computing Research Technology Platform (SCRTP) at the University of Warwick. H.K, T.J.H. and I.O. would like to acknowledge Andrew Iskauskas for crucial/ helpful discussions on statistical analysis of SEM measurements.

Author contributions

G.C.S., H.K, T.J.H., I.O. and I.J.P. conceived the research. I.J.P., F.B. and G.C.S. performed and analysed the molecular dynamics and metadynamics simulations. P.L.J., G.C.S. and J.E.S. devised applied and analysed the fluid dynamics-based mathematical model. I.O., T.J.H., H.K. and M.W.G. devised the experimental studies. I.O. and N.R. extended the methodology and
performed the fluorescence-based porosity assays. C.K.I. developed the SEM methodology and performed, together with I.O., the EM measurements. I.O., I.J.P., P.L.J., F.B., T.J.H., J.E.S., H.K. and G.C.S. contributed to writing the manuscript.

Competing interests

The authors declare no competing interests.

Additional information

Supplementary information The online version contains supplementary material available at
https://doi.org/10.1038/s42003-025-07495-0.
Correspondence and requests for materials should be addressed to Heather Knight or Gabriele C. Sosso.
Peer review information Communications Biology thanks Duoduo Wang and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Kaliya Georgieva. A peer review file is available.
Reprints and permissions information is available at http://www.nature.com/reprints
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
© The Author(s) 2025

  1. ¹Department of Biosciences and Durham Centre for Crop Improvement Technology, Durham University, Durham, UK. ²Department of Chemistry, University of Warwick, Coventry, UK. Warwick Mathematics Institute, University of Warwick, Coventry, UK. □ e-mail: p.h.knight@durham.ac.uk; g.sosso@warwick.ac.uk