ساديا سا و فكس خيارات و كدز 2008

البرازيل و ساديا لديها خسارة 4-ق على المشتقات النقد الاجنبى.

ساو باولو، 27 مارس (رويترز) - سجلت شركة الأغذية البرازيلية ساديا SDIA4.SA (SDA. N) يوم الجمعة خسارة كبيرة في الربع الرابع، عكست بذلك أرباحا عن الفترة نفسها من العام الماضي بسبب خسائر حادة في مشتقات العملات.

وبلغ صافي الخسارة 2.04 مليار ريال (890.1 مليون دولار)، مقارنة بأرباح بلغت 374.5 مليون ريال في الربع الأخير من عام 2007، وفقا لما ذكرته سادية في إيداع الأوراق المالية.

وفي عام 2008، سجلت الشركة خسارة قدرها 2.48 مليار ريال، وهي أول خسارة سنوية في تاريخها الذي يعود إلى 64 عاما.

وقد حققت ساديا نفقات مالية صافية بلغت 2.7 مليار ريال في الربع الثاني من العام، حيث ارتفعت تكلفة العقود غير القابلة للتسليم إلى الأمام والاتفاقات المستقبلية إلى الأمام وخيارات العملات. وحققت الشركة مكاسب مالية بلغت 147.5 مليون ريال في العام السابق.

& لدكو؛ مع إضعاف الحقيقي والخسائر المالية الناتجة، قمنا بتحسين إدارة المخاطر وسياسات حوكمة الشركات لدينا ونفذت مراجعة الإجراءات وهيكل إدارتنا المالية، & رديقو؛ حسبما ذكرت ساديا فى بيان.

ارتفع سعر صرف الريال البريطاني الحقيقي بنسبة 20٪ مقابل الدولار في عام 2007، وعزز 14٪ إضافية في عام 2008 حتى أوائل أغسطس، مما دفع سادية وغيرها من الشركات في البرازيل إلى المراهنة على أن العملة ستبقى على مسار قوي.

وقد تعمقت هذه الرهانات مع تعمق الاضطرابات في الأسواق العالمية وأدت المخاوف بشأن الركود العالمي إلى تدفقات رأس المال الحادة من الأسواق الناشئة.

وعلى الرغم من الخسائر الكبيرة في العملة، ارتفعت المبيعات الصافية بنسبة 15،9٪ لتصل إلى 3،06 مليار ريال، مدفوعا بزيادة في المنتجات الغذائية المجمدة والدواجن.

وانخفض سهم ساديا بنسبة 2.3 فى المئة ليصل الى 3.01 ريال يوم الجمعة، مقارنة بانخفاض مؤشر بوفيسبا المرجعي بنسبة 1.6 فى المائة. انخفض السهم بنحو 68 في المئة منذ أواخر سبتمبر، عندما كشفت لأول مرة عن الاستثمارات في سندات ليمان براذرز LEHMQ. PK ومشتقات العملات. ($ 1 = 2.292 ريس) (تقرير إلزيو باريتو؛ تحرير كارول المطران)

تأخرت جميع الاقتباسات لمدة 15 دقيقة على الأقل. انظر هنا للحصول على قائمة كاملة من التبادلات والتأخير.

تحليل مقارن للسالمونيلا الجينوم يحدد شبكة التمثيل الغذائي للنمو المتصاعد في الأمعاء الملتهبة.

يتكون جنس السالمونيلا مجموعة من مسببات الأمراض المرتبطة بأمراض تتراوح بين التهاب المعدة والأمعاء وحمى التيفوئيد. أجرينا تحليلا في السيليكو من الجينوم السالمونيلا ريانوتاتد نسبيا لتحديد التوقيعات الجينية مؤشرا على إمكانية المرض. من خلال إزالة العديد من التناقضات الشرح وعدم الدقة، عملية إعادة تعيين تحديد شبكة من 469 الجينات المشاركة في الأيض اللاهوائي المركزي، الذي كان سليما في الجينومات من مسببات الأمراض المعوية ولكن تدهور في الجينوم من مسببات الأمراض خارج الأمعاء. وشملت هذه الشبكة الكبيرة مسارات تمكن مسببات الأمراض المعوية من استخدام المغذيات المشتقة من الالتهاب وكذلك العديد من التفاعلات الكيميائية الحيوية المستخدمة لإثراء والتمييز البيوكيميائي للسيلمونيلا السيروفار. وهكذا، تحليل الجينوم المقارن يحدد شبكة التمثيل الغذائي التي توفر أدلة حول استراتيجيات اكتساب المغذيات والاستفادة التي هي سمة من مسببات الأمراض الجهاز الهضمي.

أهمية.

في حين أن بعض السيروفونات السالمونيلا تسبب الالتهابات التي لا تزال مترجمة إلى القناة الهضمية، والبعض الآخر تنتشر في جميع أنحاء الجسم. هنا، قارنا الجينوم السالمونيلا لتحديد الخصائص التي تميز الجهاز الهضمي من الممرضات إكسترينتستينال. حددنا شبكة الأيض الكبيرة التي هي وظيفية في مسببات الأمراض المعدية المعوية ولكن التحلل في مسببات الأمراض خارج الأمعاء. في حين استخدم خبراء التصنيف الصفات من هذه الشبكة تجريبيا لعدة عقود لإثراء والتمييز البيوكيميائي لسيرمون السلمونيال، تشير نتائجنا إلى أنها جزء من & # x0201c؛ خطة الأعمال & # x0201d؛ للنمو في الجهاز الهضمي الملتهبة. من خلال تحديد خصائص الشبكة الأيضية كبيرة من السالمونيلا سيروفارس المرتبطة التهاب المعدة والأمعاء، لدينا في تحليل سيليكو يوفر مخططا لاستراتيجيات محتملة للاستفادة من المواد الغذائية المشتقة من الالتهاب وحافة من الميكروبات الأمعاء المتنافسة.

المقدمة.

وكان من بين أهم الأفكار التي سعت في فجر العصر الجيني المعلومات التي عقدت داخل الجينوم الممرض. في السنوات التي تلت ذلك، ظهر تدهور الجينوم المرتفع كصفة مشتركة بين مجموعات فرعية متنوعة من البكتيريا تظهر أنماط حياة متخصصة نسبيا ومسببة للأمراض، بما في ذلك أفراد جنسيات كوكيلا، المتفطرة، السالمونيلا، الشيغيلة، و يرسينيا (1، & # x02018؛ 6) . ومع ذلك، لا تزال الروابط المحددة بين تدهور الجينوم والتغييرات الرئيسية لسلوك الممرض بعيد المنال.

كممرض نموذجي وآفة في جميع أنحاء العالم من البشر والحيوانات على حد سواء، السالمونيلا هو محور مهم من البحوث الجديدة في جوانب لا تعد ولا تحصى من المرضية، من علم وظائف الأعضاء الأساسية من البكتيريا لوظيفة النظام المضيف & # x02019 الصورة. على أساس إمكاناتها المسببة للأمراض، وغالبا ما ينقسم أفراد الأنواع السالمونيلا المعوية إلى تلك التي تسبب حمى التيفوئيد أو الحمى بارافيفويد في البشر، ويسمى التيفوئيد السالمونيلا سيروفارس، وتلك المرتبطة التهاب المعدة والأمعاء المترجمة في الأفراد المناعية، ويسمى سيروفارس السالمونيلا غير الليمفاوية. ومع ذلك، فإن الخصائص التي تميز السالمونيلا سيروفارس المرتبطة التهاب المعدة والأمعاء الموضعي من تلك التي تسبب العدوى المنتشرة لا تزال غير مفهومة.

التقدم في التسلسل إنتاجية عالية تجعل المقارنة الجينية أداة قوية على نحو متزايد لتحديد الميزات التي قد تفسر الاختلافات في إمكانات المرض من السيروفونيلا السالمونيلا. ومع ذلك، فإن عملية الشرح الجينوم يمكن أن تنتج عددا كبيرا من الأخطاء، وهي النتيجة التي تعززها الإفراط في الاعتماد على الأتمتة. وعلاوة على ذلك، الجينات المتاحة للمقارنة مشروح باستخدام أساليب مختلفة، وتسلسل تركت على نحو متزايد لم تكتمل، مما يحد من قوة تحليل الجينوم المقارن.

هنا، أجرينا إعادة ترتيب مقارن يدويا من أورثولوغس من 15 الانتهاء S. & # x000a0؛ الجينوم المعوية لتحديد التوقيعات الجينية التي تميز مسببات الأمراض التي تسبب عروض المرض المختلفة. ويشير تحليلنا إلى أن إزالة التناقضات والتعليقات التوضيحية من خلال عملية تطبيع الشرح قد عزز بشكل ملحوظ من تحليل تحليل الجينوم المقارن، مما مكننا من التعرف على بصمة جينية مخفية سابقا تميز الممرضات المرتبطة بالتهاب المعدة والأمعاء من تلك التي تسبب المرض المنتشر.

إعادة مقارنة مقارن ل 15 جينوم السالمونيلا.

وقد تم الانتهاء من خمسة عشر جينيوم S & & x000a0؛ وتشمل الجينومات المعوية، التي تضم جميع السيروفرات مع مجموعة كروموسوم بلا فجوة المتاحة من نسبي في وقت بدء هذا العمل، في التحليل (انظر الشكل & # x000a0؛ S1A في المادة التكميلية). S. & # x000a0؛ إنتيريكا سيروفار باراتيفهي B هو سلالة متعددة الأشكال تحتوي على مسببات الأمراض المرتبطة الحمى بارفاتيفويد وكذلك أعضاء متنوعة جافا، والتي ترتبط مع التهاب المعدة والأمعاء (7). و S. & # x000a0؛ تسلسل باراتيفي B الجينومية المدرجة في تحليلنا نشأت من سلالة SPB7، وهو ممثل من مجموعة متنوعة جافا. وهكذا، فإن مجموعتنا تحتوي على 5 جينومات تمثل سيروفار التيفوئيد، بما في ذلك S & & # x000a0؛ التوتري سيروفار التيفية (سلالات CT18 و Ti2)، S & # x000a0؛ باراتيفي A (سلالات أتسك 9150 و أكو 12601)، و S & x000a0؛ باراتيفهي C. الجينوم العشر المتبقية تمثل سيروفارس غير اللبية.

وكان أحد الحواجز التي واجهناها في وقت مبكر خلال تحليلنا هو أن الطرق المختلفة المستخدمة لتعليق الجينومات المتاحة، جنبا إلى جنب مع عدد كبير من عدم الدقة المكتشفة في بعض التعليقات التوضيحية، جعلت أي مقارنة مباشرة للمحتوى المتدهورة (أي المعطل افتراضيا أو المحذوف) بين الجينوم غير دقيق. وبالتالي قمنا بإجراء إعادة مقارنة مقارن لبيانات أورثولوغ من جميع الجينوم 15 (انظر الجدول & # x000a0؛ S1 في المادة التكميلية)، وحددت الحذف (انظر الجدول & # x000a0؛ S2)، وجمعت المحتوى المتدهورة في كل جينوم (انظر الجدول & # x000a0 ، S3). ولكي نعكس حالة تعطلها، سنشير إلى الموقع المسمى سابقا & # x0201c؛ بسيودوجينيس & # x0201d؛ وبدلا من ذلك تم إبطالها بشكل افتراضي بتسلسل الحمض النووي أودينغ (هسس)؛ كما المعنى الحرفي ل & # x0201c؛ الجينات الزائفة & # x0201d؛ إس & # x0201c؛ فالس جين، & # x0201d؛ كما في & # x0201c؛ بدون وظيفة، & # x0201d؛ كما أنه غالبا ما يستخدم بشكل غامض للدلالة على الجينات من الحالة الفرضية المفترضة أو التي تم التحقق من صحتها، نقترح أن يكون استخدامها محجوزا لوضع العلامات حيث فقدت كل وظيفة معروفة أثبتت تجريبيا (على سبيل المثال، الجين الكاذب فيب من S. تيفهي [8] ).

كان من الممكن لأتمتة جزء فقط من عملية إعادة التصويت، الأمر الذي جعل هذه المهمة تستغرق وقتا طويلا. ومع ذلك، تم التحقق من ضرورة إجراء هذا التحليل المرهق في سيليكو من خلال تحديد التغيرات الملحوظة في المحتوى المتدهورة لكل جينوم (انظر الجدول & # x000a0؛ S4 في المادة التكميلية). على سبيل المثال، إعادة تعييننا من 15 S. & # x000a0؛ الجينوم المعوية تحديد ما مجموعه 1،004 هكس جديدة، في حين أن ما مجموعه 471 الإدخالات، التي كانت مشروح كما & # x0201c؛ الكاذبة & # x0201d؛ سابقا، تم العثور على أنها تسلسل الحمض النووي الترميز افتراضية سليمة (كدز).

يميز التوقيع الجيني اثنين من السالمونيلا الباثوفار.

والمثير للدهشة أن تحليلنا لمؤشر S & O x000a0 المعاد قياسه نسبيا؛ لم يقدم جينومات المعوية دعما مقنعا للتصنيف في سيروفار التيفويدية وغير الليمفاوية. تدهور ثلاثة جينات فقط، فوي، فليب، و STM4065، كانت فريدة من نوعها في الحاضر في جميع سيروفارس التيفية التي تم تحليلها (انظر الجدول & # x000a0؛ S3 في المادة التكميلية). وعلاوة على ذلك، كان تدهور المجموعة الجينية وكا، الذي يشفر حمض الحيوي الكولاني، شائعة وفريدة من نوعها لالجينوم من السيروفار التيفية.

ومع ذلك، اقترح تحليل المحتوى المتدهورة في كل جينوم أن S. & # x000a0؛ سيروفارس المعوية يمكن تقسيمها إلى مجموعة واحدة تحمل عدد قليل من هكس (في المتوسط ​​66 هكس في الجينوم) ومجموعة ثانية مع عدد كبير من هكس (في المتوسط ​​246 هدس لكل جينوم) (انظر الشكل & # x000a0؛ S1B والجدول & # x000a0؛ S4 في المادة التكميلية). المجموعة الأخيرة، والتي سوف نشير إليها باسم & # x0201c؛ باثوفار إكسترينتستينال، & # x0201d؛ تم تشكيلها من قبل سيروفارس المضيف تكييفها حصرا مع العدوى المنتشرة في خزاناتها البشرية أو الحيوانية منها. الجينومات التي تحمل توقيع هك من باثوفار إكسترينتستينال شملت تلك S. S. إنتيرنيكا سيروفار & # x000a0؛ تشوليرايسويس، الذي يرتبط مع تجرثم الدم في الخنازير، S. إنتيريك سيروفار دبلن، وهو سبب تجرثم الدم في الماشية، S. إنتيرنيك سيروفار & # x000a0؛ غاليناروم ، وهو العامل المسبب لتيفوئد الطيور في الدواجن، وكذلك جميع السلوفونات التيفية السالمونيلا المدرجة في تحليلنا (أي S & & # x000a0؛ باراتيفهي A، S. & # x000a0؛ باراتيفي C، و S. تيفي). الجينوم التي تتميز بانخفاض عدد هكس تنتمي إلى S. سيركوفار المعوية أغونا، S. إنتيرك سيروفار & # x000a0؛ إنتيريتيديس، S. إنتيريكا سيروفار هايدلبرغ، S. إنتيريكا سيروفار نيوبورت، S. إنتيريكا سيروفار ششوارزنغروند، S. إنتيريكا سيروفار & # x000a0؛ تيفيموريوم، و S. & # x000a0؛ باراتيفي B. سوف نشير إلى المجموعة الأخيرة باسم & # x0201c؛ باثوفار الجهاز الهضمي، & # x0201d؛ لأن جميع أعضائها يحمل مجموعة واسعة من المضيفين وترتبط مع التهاب المعدة والأمعاء في بعض الأنواع المضيف على الأقل. وتجدر الإشارة إلى أن بعض أعضاء باثوفار الجهاز الهضمي هي أيضا قادرة على التسبب التهابات خارج الجهاز الهضمي في بعض المضيفين. على سبيل المثال، S & # x000a0؛ يرتبط تيفيموريوم مع تجرثم الدم في الفئران. ومع ذلك، يسبب مسببات المرض التهاب المعدة والأمعاء الموضعي في الماشية وفي البشر المناعي. وهكذا، فإننا نشير إلى هذه المجموعة باسم باثوفار الجهاز الهضمي، لأن القدرة على التسبب في التهاب المعدة والأمعاء في بعض الأنواع المضيف على الأقل يفترض أن يضع الجينات اللازمة لهذا النمط من الحياة تحت الاختيار.

تم الكشف عن العديد من التوقيعات الجينومية التي تدعم التمييز بين باثوفار الجهاز الهضمي و باثوفار إكسترينتستينال في تحليلنا. تحليل كدز التي كانت في كثير من الأحيان متدهورة (n & # x02265؛ 4) في أعضاء مجموعة واحدة ولكن نادرا (n & # x02264؛ 1) في أعضاء آخرين دعم تصنيف إلى اثنين من باثوفارس ولكن قدمت رؤى وظيفية قليلة (انظر الجدول و # x000a0؛ S5 في المادة التكميلية). كشف تحليل الجينات المتورطة في الفوعة أن الجينومات التي تمثل باثوفار إكستراستينستينال عرضت حالات أكثر من الجينات المتدهورة ترميز النوع الثالث يفرز البروتينات المؤثرات، أدهسينس فيمبريال، والوظائف المتعلقة الحركة والكيموتاكسيس من الجينومات التي تمثل باثوفار الجهاز الهضمي (انظر الجدول & # x000a0؛ S6 والشكل & # x000a0؛ S2)، وهو ما يتسق مع تقرير سابق (6). فيمبرياي، والحركية، والكيميائية مطلوبة لاستعمار الأمعاء (9، & # x02018؛ 11) ولكن لا يمكن الاستغناء عن البقاء على قيد الحياة في الأنسجة المضيف (12، 13)، والتي قد تفسر لماذا يتم الحفاظ على هذه الوظائف في باثوفار الجهاز الهضمي ولكن تخضع للتدهور في ال التعريف، إكستراينتستينال، باثوفار.

وكانت النتيجة الأكثر لفتا لدينا في تحليل السيليكو من الجينوم السالمونيلا ريانوتاتد نسبيا تحديد شبكة التمثيل الغذائي كبيرة تتألف من 469 كدز، 167 منها كانت متدهورة بشكل فريد في واحد أو أكثر من الجينومات في باثوفار إكسترينتستينال (الشكل & # x000a0؛ 1 ؛ انظر أيضا الجدول & # x000a0؛ S7 في المادة التكميلية). كان العدد الإجمالي لل هكس و كدز المحذوفة التي تنتمي إلى هذه الشبكة الأيضية، لا عد حالات مكررة من سلالات تنتمي إلى نفس سيروفار، 224 لجميع الجينومات التي تمثل باثوفار إكسترينتستينال، مقارنة فقط 13 لجميع الجينومات التي تمثل باثوفار الجهاز الهضمي (نسبة من 17.23). وكشف التحليل الإحصائي أن نسبة 17.23 هي حوالي 9 الانحرافات المعيارية بعيدا عن متوسط ​​النسبة التي تم الحصول عليها عندما يتم تحديد المحتوى المتدهورة للمجموعات بالسكان عشوائيا من 469 كدز من كل جينوم (P.

في حين أن التدهور الزائد إحصائيا من الجينات الأيضية التي تم تحديدها هنا قدمت دعما مقنعا للتمييز بين باثوفار إكستراستينستينال من باثوفار الجهاز الهضمي، لم يكن هذا التصنيف مدعومة من قبل الشروح الجينوم السابقة. باستخدام التعليقات التوضيحية المنشورة، كشف تحليل 469 كدز التي تنتمي إلى الشبكة الأيضية المبينة في الشكل & # x000a0؛ 1 169 كدز المتدهورة في باثوفار إكسترينتستينال مقارنة ب 46 في باثوفار الجهاز الهضمي. لم تكن النسبة الناتجة 3.67 مختلفة اختلافا كبيرا (P = 0.17) عن النسبة التي لوحظت في مجموعات مختارة عشوائيا من 469 كدز من كل جينوم، وهو ما يفسر لماذا لم يحدد التحليل السابق لهذه الجينوم السالمونيلا هذه الشبكة الأيضية الكبيرة (14). وهكذا، حتى الآن، حقيقة أن شبكة من 469 كدز تشارك في الأيض اللاهوائي المركزي هو التحطيم في الجينوم من باثوفار إكسترينتستينال ظلت مخفية وراء الضوضاء الإحصائية الناتجة عن التناقضات وعدم الدقة في الشروح الجينوم السابقة.

شبكة الأيض الكبيرة التي تحتوي على وظائف لاستخدام المواد الغذائية المستمدة من الالتهاب هو مهينة في باثوفار إكسترينتستينال.

شبكة التمثيل الغذائي الناشئة من تحليلنا يتضمن العديد من الوظائف التي سبق أن أظهرت أن تكون مهمة للنمو اللاهوائي في تجويف الأمعاء أثناء التهاب المعدة والأمعاء. S & & # x000a0؛ تيفيموريوم، وهو عضو في باثوفار الجهاز الهضمي، ويستخدم نوعه إفراز النظم الثالثة المشفرة بواسطة السالمونيلا المسببة للأمراض الجزيرة 1 (SPI1) و SPI2 لتحريك التهاب الأمعاء الحاد (15). وهناك نتيجة ثانوية للرد على استجابة المضيف التهابات هو توليد مستقبلات الإلكترون محطة نترات وتيتراثيونات، وجود منها يعزز النمو اللمعي للممرض عن طريق التنفس اللاهوائي (16، 17). وحدد تحليلنا هذه المسارات جنبا إلى جنب مع العديد من الوظائف الإضافية المتعلقة بالتنفس اللاهوائي، والتي تنطوي على نقل الإلكترونات من المتبرع، مثل الفورمات أو اللاكتات أو الهيدروجين (H2)، من خلال تجمع كينون إلى متقبل، مثل النترات، تيتراثيونات، النتريت، أكاسيد S، N - أكاسيد، أكسيد النيتريك، ثيوسلفات، أو كبريتات (الشكل & # x000a0؛ 1). فورمات، اللاكتات، والهيدروجين هي منتجات التخمير نهاية ولدت من قبل المجتمعات الميكروبية اللاهوائية تلزم يسكن الأمعاء البعيدة (18، 19)، والهيدروجين المستمدة من الجراثيم وقد ثبت مؤخرا أن تغذي نمو S & # x000a0؛ تيفيموريوم في التجويف الأمعاء الكبيرة (20).

وجود مستقبلات بديلة للإلكترون، مثل تيتراثيونات، يمكن S & & # x000a0؛ تيفيموريوم أن ينمو على مصادر الكربون غير القابلة للتخليق الأخرى، مثل الإيثانولامين، الذي ينتج عن تدهور الميكروبات من فوسفهاتيديليثانولامين وفيرة فوسفاتيبيد في الأمعاء البعيدة (21). أظهرت الجينومات التي تمثل الباثوفار خارج الأمعاء تدهور كدز تشارك في استخدام الايثانولامين (جينات اليوت)، وكذلك في التركيب الحيوي للفيتامين ب 12 (الجينات المركزة و الكوب)، عامل مساعد المنتجة تحت ظروف اللاهوائية، وهو مطلوب للاستخدام الإيثانولامين (22) ) (الشكل & # x000a0؛ 2).

فيتامين ب 12 ضروري أيضا لاستخدام 1،2-بروبانديول، كاتابوليت التي تنتجها الميكروبات تخمر فوكوس أو رامنوس. التعبير عن S & # x000a0؛ البروتينات تيفيموريوم تشارك في هدم السكر في زيادة تجويف الأمعاء في نموذج التهاب القولون الماوس (23). وعلاوة على ذلك، المجتمعات من البكتيريا اللاهوائية الملزمة في الأمعاء البعيدة تحرير المضيف المخاطية المستمدة من السكريات الأحادية، مثل فوكوس، مما يؤدي إلى زيادة التعبير عن S & # x000a0؛ جينات تيفيموريوم تشارك في تدهور فوكوس (فوك الجينات) ومنتج التخمير لها 1،2-بروبانديول (بدو الجينات) في تجويف الأمعاء من الفئران مونواسوسياتد مع باكتيرويدس ثيتايوتوميكرون مقارنة مع الفئران جيرمفري (24). وحدد تحليلنا تدهور كبير في باثوفار خارج الشبكة عبر شبكة كبيرة من الجينات المشاركة في امتصاص وتدمير هرمونات أحادية مختلفة، والتي شملت الجينات فوك و بدو (الشكل & # x000a0؛ 1).

إلى جانب المسارات التي ظهرت سابقا في الدراسات على نمو اللمعية من S & # x000a0؛ تيفيموريوم أثناء التهاب القولون، حددت شبكتنا العديد من الوظائف الجديدة التي من المرجح أن تسهم في عملية التمثيل الغذائي اللاهوائي المركزي من باثوفار الجهاز الهضمي. على سبيل المثال، تدهور كدز تشارك في اللاهوائية & # x003b2؛ - أكسدة الأحماض الدهنية كانت ممثلة تمثيلا زائدا في الجينومات التي تمثل باثوفار إكسترينتستينال. هذا المسار، الذي يختلف عن الهوائية و # x003b2؛ - أكسدة مسار لتدهور الأحماض الدهنية، ويتم ترميز من قبل يديفو، يديكرست، وجينات فادهيك ويتطلب وجود بديل الإلكترون متقبل، مثل نترات، S - أكاسيد، أو N - أكاسيد (23). ومن المثير للاهتمام أن الأحماض الدهنية القصيرة السلسلة تتراكم في تجويف الأمعاء البعيدة عندما تنهار مجتمعات البكتيريا اللاهوائية الملوثة وتخمر الكربوهيدرات المعقدة، في حين يتم توليد النترات في هذه البيئة كمنتج ثانوي لاستجابة المضادات الالتهابية (17)، والتي عند ظهور S & # x000a0؛ تيفيموريوم ينشر أنظمة إفراز النوع الثالث المشفرة بواسطة SPI1 و SPI2 (15).

أظهرت جميع الجينومات السالمونيلا تدهورا ضئيلا جدا في كدز تشارك في وظائف التمثيل الغذائي المركزي المطلوبة في ظل الظروف الهوائية، ويرجع ذلك على الأرجح لأن هذه الصفات ضرورية للنمو البكتيري في الأنسجة المضيفة (25). على سبيل المثال، ظلت الجينات المتورطة في دورة غليوكسيلات، وهي البديل اللاهوائي لدورة حمض التيربوكسيليك الهوائية، سليمة، لأن وظائفها مطلوبة أيضا للإصدار الهوائي لهذا المسار. ومع ذلك، كان تدهور كدز تشارك في امتصاص المركبات من البيئة التي يمكن تجديد الوسطاء في دورة غليوكسيلات، مثل سيترات، طرطرات، تريكارباليلات، سيرين، و أسبارتات تمثيلا زائدا في الجينومات التي تمثل باثوفار إكسترينتستينال (الشكل & # x000a0 ؛ 1). وعلاوة على ذلك، فإن كدز المطلوبة لتفاعلات التسمم التي تملأ الفجوة بين 2-أوكسوغلوتارات والسكسينات في دورة غليوكسيلات اللاهوائية كانت عادة المتدهورة في الجينومات التي تمثل باثوفار إكسترينتستينال. هذه التفاعلات الابتنائية ليست مطلوبة في ظل الظروف الهوائية، لأن سوكا و سوسب تحويل 2-أوكسوغلوتارات إلى سوتسينيل-أنزيم A (كوا) ضمن دورة حمض ثلاثي الكربوكسيل.

وأخيرا، أظهرت الجينومات التي تمثل باثوفار إكسترينتستينال تدهور المنظمين لمجموعة متنوعة من العمليات اللاهوائية، بما في ذلك التنفس اللاهوائي (ناربق، نور، تورستر، ترس)، وما يترتب على ذلك من تدهور لاهوائي من منتجات التخمير والأحماض الدهنية (لدر، بوكر، بربر، و يديب )، هدم الكربوهيدرات (دغور، غالس، ربسر، رار، أوبك، ياج)، والوظائف المتعلقة بدورة غليوكسيلات اللاهوائية (أسيك، دكوس، دبيب) (الشكل & # x000a0؛ 1 و 2).

نقاش.

وشملت الشبكة الأيضية الكبيرة المحددة في تحليلنا (الشكل & # x000a0؛ 1) العديد من علماء التفاعلات الكيميائية الحيوية والمختبرات السريرية تستخدم لعزل وتمييز السيلمونيلا السيلوفار. على سبيل المثال، كان النمو في مرق تحتوي على تيتراثيونات قيد الاستخدام منذ عام 1923 كوسيلة لإثراء للسالمونيلا السيروفار في العينات التي تحتوي على الميكروبات الأخرى (26). وتتبع هذه الثقافة الإثراء الأولي عن طريق الكشف عن إنتاج كبريتيد على الحديد أو البزموت التي تحتوي على أجار انتقائية، مثل الثلاثي السكر أجار الحديد مائلة وضعت في عام 1917 (27) أو لوحات البزموت الكبريت أجار وضعت في عام 1923 (28). في حين تم استخدام هذه الصفات الأيضية تجريبيا لعقود عديدة لعزل السالمونيلا السيروفار، تحليلنا تشير إلى أنها جزء من شبكة الأيض الكبيرة التي تعرف باثوفار الجهاز الهضمي. منذ الغالبية العظمى من أكثر من 2500 ق. & # x000a0؛ ويرتبط سيروفارس المعوية مع التهاب المعدة والأمعاء في البشر مناعيا، قد يكون من المستغرب أن هذه الوظائف غالبا ما تعتبر سمة من سمات كامل S & # x000a0؛ الأنواع المعوية، على الرغم من أنها مهينة في الجينوم من عدد قليل من المتخصصين الذين ينتمون إلى باثوفار خارج المعوية.

يتم استخدام التدهور في باثوفار إكسترينتستينال من الوظائف التي تنطوي على الأيض المركزي اللاهوائي (الشكل & # x000a0؛ 1) تجريبيا للتمييز الممرضات المرتبطة الحمى باراثيفويد من الكائنات ذات الصلة وثيقة التي لا يمكن أن تكون متباينة عن طريق الأنماط المصلية ولكن تسبب التهاب المعدة والأمعاء في البشر. مثال واحد هو S & & # x000a0؛ باراتيفي B متنوعة جافا، الممرض المرتبطة التهاب المعدة والأمعاء البشري، الذي له نفس الصيغة مستضد (1،4 [5]، 0.12: b: 1،2) كما S & & # x000a0؛ باراتيفي B، وهو سبب الحمى بارفاتيفويد. يتم استخدام القدرة على تخمير طرطرات تجريبيا للتمييز بين هذه مسببات الأمراض بيوشيميكالي (7). في حين أن S. & # x000a0؛ باراتيفي B متنوعة يمكن لعزلات جافا تخمير الطرطرات، وهذا المسار الذي يسهم في الشبكة الأيضية المحددة في تحليلنا تعطلت بواسطة انتقال النوكليوتيدات من G إلى A ضمن كود بدء أتغ STM3356 في S & x000a0؛ باراتيفي ب العزلات من المرضى الذين يعانون من حمى بارفاتيفويد (29). المثال الثاني هو S. & # x000a0؛ إنتيريكا سيروفار سينداي، وهو سبب حمى باراتيفويد، الذي له نفس الصيغة المستضدية (1،9،12: a: 1،5) كما S & & # x000a0؛ إنتيريكا سيروفار ميامي، وهو سبب التهاب المعدة والأمعاء البشري. ويمكن تمييز كل من مسببات الأمراض كيماوية حيوية، لأن عزلات S. ميامي يمكن أن تخمر سترات، في حين أن عزلات S. سينداي سلبية لهذه التفاعلات داخل الأيض المركزي اللاهوائي (30).

من وجهة نظر سيروفارس بين S & # x000a0؛ المعوية، وتحليلنا من الجينومات المعاد توجيهها نسبيا يمثل أوسع دراسة متعمقة من تدهور السالمونيلا الجينوم حتى الآن. في هذا الصدد، فإن الأصل أحادي الطيف وتشابه عالية من S. العزلات تيفي (31)، إلى جانب التاريخ المتعدد المضيف معزولة من باثوفار إكسترينتستينال (انظر الشكل & # x000a0؛ S1 في المواد التكميلية) (32) و إدراجنا مجموعة واسعة مماثلة من سيروفارز الجهاز الهضمي (انظر الشكل و # x000a0؛ S1)، تشير إلى أن مجموعة البيانات لدينا هي متنوعة بشكل مناسب. هذه الاعتبارات، جنبا إلى جنب مع احتمال ضعيف للغاية أن تدهور الأيض اللاهوائي المركزي نشأت عشوائيا في جميع أعضاء تحليلها من باثوفار إكستراستينستينال، فضلا عن احتمال غير مماثل أن الفرق في تدهور المذكور بين باثوفارس هو قطعة أثرية الناشئة عن جينوم 15 محددة نحن تحليلها، تعطينا الثقة بأن ملاحظاتنا سوف تصدق كما أكثر سلالات و سيروفارس التسلسل؛ في الواقع، فإننا نتوقع أن توسیع عدد الجینومات التي تم تحلیلھا سوف یجلب الأنماط التحلیلیة الأکثر دقیقة، التي یمکن أن تکون مضیفة للمضیفة إلی البروز.

ومع ذلك، هناك العديد من أشكال تغيير الجينوم التي هي، في الوقت الحاضر، أكثر صعوبة لفرض آثار من خلال في تحليل سيليكو وحدها. وتشمل هذه الحالات تحديد والأدوار التكيفية للأليلات الجديدة هبومورفيك الناشئة عن الطفرات الصاخبة (على سبيل المثال، أليل E211 من بميرا في S & X x000a0؛ باراتيفي ب) (33)، ونتائج الطفرة داخل العناصر التنظيمية النشطة - رابطة الدول المستقلة، قطبية إندلس تقع داخل أوبيرونات معروفة أو مفترضة، وتأثير اكتساب المنظم من خلال نقل الجينات الأفقي (على سبيل المثال، التعديلات التنظيمية التي أدلى بها تفا من S. تيفي) (34، 35). وعلى هذه الجبهة، فإن التحليل التجريبي ضروري لتيسير تحديدها وترشيدها. وتظهر ضرورة التحليل التجريبي بشكل واضح من خلال مثال الجين فيب الذي يشفر منظما لسلسلة مستضدات O-أنتيجن طويلة جدا (& # x0003e؛ 100 وحدة متكررة) (36)، وهي بنية سطحية تمنح مقاومة الصفراء في S. & # x000a0؛ تيفيموريوم (37). في الجينوم S. تيفي، وتعطل فيب إطار القراءة المفتوحة عن طريق كودون وقف (2)، مما أدى إلى فقدان سلاسل طويلة جدا O - مستضد (8). ومن المثير للاهتمام، وهذا فقدان سلاسل طويلة جدا O - المستضد يعظم التهرب المناعي بوساطة الفاسدة المرتبطة (السادس) المحفظة السكاريد من S. تيفي (38). وبالتالي، فإن عواقب تكوين الكاذب يمكن أن تكون معقدة، مما يدل على الحاجة لمتابعة في دراسات السيليكو مع تحليل تجريبي.

ومع ذلك، فإن وضع التوقيعات الجينومية التحللية التي اكتشفناها في تحليل السيليكو لمؤشر S & x x000a0 المعاد قياسه نسبيا؛ الجينومات المعوية في سياق الجسم القائم من العمل على بيولوجيا هذه العوامل الممرضة تدعم نموذجا يميز اثنين من باثوفارس، كل استغلال مختلف مكانة المضيف لنقل. يستخدم أعضاء باثوفار الجهاز الهضمي عوامل الفوعة الخاصة بهم للحث بسرعة على التهاب الأمعاء الحاد (15) واستغلال التغيرات التي تلت ذلك في البيئة من خلال تعزيز نموها اللمعية باستخدام شبكة الأيض الكبيرة تشارك في الأيض اللاهوائي المركزي (الشكل & # x000a0؛ 1 ) (11، 16، 17، 21، 24). وتزهر اللمعة الناتجة عن أعضاء باثوفار الجهاز الهضمي يعزز انتقالها عن طريق الفم البراز (39).

في المقابل، S. تيفي، وهو عضو في باثوفار إكسترينتستينال، يوقف في البداية التهاب الأمعاء (38، 40، 41) ويسبب عدوى نشرها تعرف باسم حمى التيفوئيد. وهناك جزء صغير) حوالي 4٪ (من األفراد الذين يتعافون من حمى التيفوئيد يطورون نقل املرارة املزمن وهم املخزون الرئيسي لنقل حمى التيفوئيد) 42 (. في حين أن أعضاء آخرين من باثوفار إكسترينتستينال أيضا تسبب العدوى المنتشرة، وبعض استغلال مختلف الأجهزة للإرسال، مثل المبايض في حالة S. & # x000a0؛ غاليناروم (43) أو الضرع في حالة S. دبلن (44) . ومع ذلك، في كل حالة، ويسهل انتقال الكائن الحي و 's عن طريق نشر تليها استمرار المزمن في الأنسجة المضيفة، والبيئة ميكرويروبيك (25)، مما يجعل الجينات المطلوبة للنمو اللاهوائي في الأمعاء البعيدة للاستغناء عن باثوفار إكسترينتستينال. ويظهر تحليلنا أن التدهور الناتج من الوظائف التي تنطوي على الأيض اللاهوائي المركزي هو تجربة الطبيعة التي أنتجت بصمة جينية بارزة مميزة من الجينومات التي تمثل باثوفار إكسترينتستينال. من خلال تحديد وظائف مهينة في الجينومات من باثوفار إكسترينتستينال، حددت دراستنا شبكة الأيض كبيرة من المرجح أن يجسد & # x0201c؛ استراتيجية الفوز & # x0201d؛ التي يعمل بها أعضاء باثوفار الجهاز الهضمي إلى حافة خارج الميكروبات المتنافسة في تجويف الأمعاء الملتهبة، وبالتالي تعزيز انتقالها.

المواد والأساليب.

إعادة المقارنات المقارنة.

لكل الجينوم الذي تم تحليله (انظر القائمة في أعلى الجدول & # x000a0؛ S1 في المادة التكميلية) & # x000a0؛ (2، 6، 14، 45، & # x02018؛ 50)، جمعنا كل كدز و كادس الزائفة المعلومات عن طريق تحليل سجلات البنك الوطني نسبي. ثم حصلنا على قاعدة بيانات ونيبروت كنولدجيباس (51) لهذه المواقع عن طريق الإحالة المرجعية جينريز إنتريز (52) وتحليلها لأسماء الجينات، والتعليقات الوظيفية، و كوغ المرتبطة بها (53)، بفام (54)، و تيغرفام (55) المجالات البروتين . لتطبيع الشروح أورثولوغ، اتخذنا واحدة كدز في وقت واحد من الفهرس كمرجع وتقع أورثولوغس في الجينومات الأخرى، المسببة للعمى الخيارات المرجعية الأولية لوظيفة الجينات وانحيازه إلى الأقل المتدهورة الجينومات المنسوجة يدويا (S & & # x000a0 ؛ تيفيموريوم LT2، S. & # x000a0؛ إنتيريتيديس P125109).

للتعليق على أورثولوغس، كتبنا مخطوطات مخصصة لتحليل تحالفات التسلسل المرجعية التي تم إجراؤها على الجينوم الموضوع مع بلاستن و تبلاستن عبر نسبي و # x02019؛ s واجهة برمجة تطبيقات الويب (أبي) (56). وباختصار، تحليلنا النصي وتحليل نتائج بلاست، أكدنا يدويا محاذاة دقيقة السياق، ومن ثم سيناريو الإحداثيات المتكاملة وتسلسل المعلومات من كلا أساليب بلاست لتحديد حدود الجين المرجعي في الجينوم الموضوع؛ إذا لم يتم العثور على بداية الانحياز أو وقف كودون، ونحن تفتيشها يدويا المنطقة. ثم حل البرنامج النصي المحاذاة لإدراج، حذف، كودونات توقف سابق لأوانه، فرامشيفتس، والتغييرات على كودون البداية. نحن نحدد هك ليكون موضع أورثولوغوس مع & # x02265؛ 10 كودونات تعطلت من قبل الطفرات المذكورة أعلاه بالنسبة إلى كدز مرجع. كان محاذاة في نفس السياق الجيني مع & # x02265؛ 90٪ من الأحماض الأمينية الهوية، باستثناء الثغرات و اقتطاع، وكان لدينا قطع الأولي ل أورثولوغي. منحت أن أي قطع من هذا القبيل تعسفية، افترضنا أن أكبر تغييرات إطار القراءة المفتوحة إلى كدز مشابهة للغاية من المرجح أن إشارة وظيفة تعطيل. وبالتالي، تم اختيار قطع قطعنا لتجنب الضوضاء في شكل أحداث أصغر، يحتمل أن تكون غير معطلة (على سبيل المثال، اقتطاع من كودون واحد). في هذا الصدد، لدينا قطع قطع تعطيل فعال أقل من أو يساوي جميع القطع السابقة بين الجينوم تحليلها، كما يتضح من أكثر حالتين في الجينوم (انظر الجدول & # x000a0؛ S4 في المواد التكميلية، & # x0201c؛ الآن غير واضح & # x0201d؛ العمود) من المكالمات السابقة للجينات الزائفة التي تنطوي على انقطاع محتمل لا يفي بقطع المقاس. ومع ذلك، يتم وضع علامة على جميع الأحداث الفرعية قطع & # x0201c؛ غير واضح & # x0201d؛ في الجداول التكميلية ينبغي للقارئ الرغبة في النظر فيها.

بعد ذلك، إذا لم يكن التعليق التوضيحي للأغلبية مطابقا للمرجع، فقد قمنا بالتحقيق في المرجع وتحويله إلى التعليق التوضيحي أورثولوغ & # x02019 إذا كان ذلك مناسبا. Prior to selecting a new reference, our script removed any locus tags from the index that were associated with identified orthologs. Table S1 in the supplemental material contains data collected on each ortholog, with the genome of LT2 serving as a scaffold for ordering entries and with episomal data placed at the end of the list. The Table S1 legend describes the data and provides associated cutoffs.

To preclude analyzing potentially overannotated genome content, we discarded CDSs ≤75 codons from the potential reference index unless they bore an annotated function, informative homology, or a protein domain. References found within prophage or mobile genetic elements were compared only for orthologs with similar regions located in the same genomic context. As the expression of integrases and transposition-related genes is not known to immediately impact the pathobiology of Salmonella serovars, we did not meticulously investigate these entries or mark them as intact or disrupted; we identified these loci using the ISFinder database (57) and CD-Search (58). Regarding previously annotated pseudo-CDSs that did not associate with intact references, we checked for disruptions relative to nonorthologous references and then checked for orthologs, discarding small fragments and loci that were disrupted in all analyzed strains, as their differential role in genome degradation was unclear at this juncture.

Deletions and truncations.

To identify disruptive lesions, we located remnants of reference loci from Table S1 in the supplemental material and of RNA genes as an indicator that a gene or region was present and subsequently truncated or deleted. Table S2 in the supplemental material contains a list of alignment gaps within, and extending outside, at least one locus and that we propose to be disruptive (see Table S2 for definitions and cutoffs; Table S1 data contains intragenic indels). In brief, we wrote scripts and used manual curation to systematically compare partially overlapping segments of S . Typhimurium LT2 against all other analyzed genomes, utilizing the megablast algorithm of blastn via the BLAST Web API (56) with a high-scoring alignment pair cutoff of 80% identity, and then catalogued alignment gaps residing within the same genomic context. We then compared regions in the same context that were missing from LT2 and filtered out highly mosaic regions and dissimilar prophage insertions in the same context from further examination. Our script identified gap intersections with reference locus coordinates and calculated disruptions, which we then manually curated and swapped with other regions to serve as a reference when the original reference appeared to be affected, updating Table S1 references as necessary.

We marked missing regions without a flanking remnant as absent. If an absent region from one strain resided completely within a proposed deletion in another strain, we marked that section of the deletion as absent. When reference DNA was plausibly not present (e. g., mobile element insertion) prior to a proposed deletion having occurred, or when stepwise intermediate genotypes were unavailable to resolve multiple instances having occurred, we marked the region as absent and marked the disrupted border gene(s) as truncated.

CDS groupings.

To identify pathways involved in central anaerobic metabolism, we examined primary literature, associated entries in the Kyoto Encyclopedia of Genes and Genomes (59), and Escherichia coli K-12 ortholog entries in the BioCyc database (60). To index genes involved in other aspects of pathogenesis, we used protein domains to identify chaperone-usher fimbrial gene clusters (61), obtained the identities of type III secretion system effectors primarily from reference 62, and utilized the S . Typhimurium FlhDC regulon (63) to populate our list of motility and chemotaxis CDSs.

To calculate the probability of the observed extraintestinal-to-gastrointestinal pathovar ratio of total degradation in the central anaerobic metabolism group (3.67 before reannotation, 17.23 after) having occurred at random, we generated 250 random groups of 469 reference loci present or once present in ≥10 of the analyzed genomes; multiple hits for a reference locus within a serovar were tallied only once. From this data set, we log-transformed the ratios and computed the mean (0.482) and standard deviation (0.088) of the random group ratios and then used a quantile-quantile plot to confirm that the log-transformed random ratios closely fit a normal distribution (trendline of y = 0.9945 x + 6 × 10 −16 , R 2 = 0.9902). With these values, we computed the z  scores (before = 0.945, after = 8.598) and one-tailed P  values (0.172,

0) for the log-transformed observed ratios (0.565, 1.236).

SUPPLEMENTAL MATERIAL.

Fifteen genomes representing 13 S. enterica serovars selected for analysis. Genomes representing the extraintestinal pathovar are indicated in blue font. Panel A is an unrooted phenogram illustrating the phylogenetic relatedness of the selected genomes. From each genome, we concatamerized, in the same order, the nucleotide sequences of 2,651 intact CDS orthologs (highlighted in the “Index” column of Table S1 in the supplemental material) that are conserved across all analyzed genomes. We then aligned the concatamers with MUSCLE 3.8.31 using the “refinew” parameter and analyzed the alignment with the phylogeny inference package (PHYLIP 3.695). To generate the unrooted phenogram, we used DNADIST, NEIGHBOR, and DRAWTREE with default settings; to bootstrap the alignment, we used SEQBOOT, DNADIST, and NEIGHBOR, each set to 1,000 replicates, with random seed “123” when needed, followed by CONSENSE with default settings. All nodes are supported by bootstrap values of >77%. (B) The graph shows the number of hypothetically disrupted CDSs (HDCs) detected in each bacterial genome (see Table S4 in the supplemental material). تحميل.

Degradation of pathogenesis-related CDS groupings. Panel A displays the names of potentially disrupted or deleted CDSs involved in motility and chemotaxis within each genome analyzed. Panel B contains all genes in each genome that encode effectors secreted by the Salmonella pathogenicity island-2 type III secretion system. Panel C provides the names of all chaperone-usher gene clusters in each genome. A white box indicates that the gene or gene cluster is unaffected, and a blue box indicates that a potential disruption or deletion of the locus has occurred. تحميل.

Table S2.

Deletions and truncations.

Disruptions and status changes.

Table S5.

Commonly disrupted/deleted CDSs.

CDS lists and tallies for groups.

CDSs from central anaerobic metabolism model.

شكر وتقدير.

We are grateful to Renée M. Tsolis for helpful suggestions on the manuscript.

We acknowledge support by Public Health Service grants AI044170 ; and AI096528 to A. J.B. and by the Floyd and Mary Schwall Fellowship in Medical Research and Public Health Training Grant AI060555 to S.-P. N.

Citation Nuccio S, Bäumler AJ. 2018. Comparative analysis of Salmonella genomes identifies a metabolic network for escalating growth in the inflamed gut. mBio 5(2):e00929-14. doi:10.1128/mBio.00929-14.

Design of a Customized Multipurpose Nano-Enabled Implantable System for In-Vivo Theranostics.

The first part of this paper reviews the current development and key issues on implantable multi-sensor devices for in vivo theranostics. Afterwards, the authors propose an innovative biomedical multisensory system for in vivo biomarker monitoring that could be suitable for customized theranostics applications. At this point, findings suggest that cross-cutting Key Enabling Technologies (KETs) could improve the overall performance of the system given that the convergence of technologies in nanotechnology, biotechnology, micro&nanoelectronics and advanced materials permit the development of new medical devices of small dimensions, using biocompatible materials, and embedding reliable and targeted biosensors, high speed data communication, and even energy autonomy. Therefore, this article deals with new research and market challenges of implantable sensor devices, from the point of view of the pervasive system, and time-to-market. The remote clinical monitoring approach introduced in this paper could be based on an array of biosensors to extract information from the patient. A key contribution of the authors is that the general architecture introduced in this paper would require minor modifications for the final customized bio-implantable medical device.

1 المقدمة.

The current interaction between medicine and technology permits the development of new diagnostic devices to detect or monitor pathogens, ions, diseases, etc. Doubtless, the integration of rapid advances in areas such as microelectronics, microfluidics, microsensors and biocompatible materials entails the availability of implantable biodevices for continuous monitoring or event detectors that carry out faster and cheaper clinical tasks than when these are done by standard methods. Implantable devices have already been used in millions of patients [1]. Benefits of these approaches include improved care and quality of life [2]. Implantable sensor networks can facilitate an early detection of emergency conditions and diseases in patients at risk, [3] comprising physical, physiological, psychological, cognitive, and behavioral processes [4], by reaching inaccessible environments in a reduced response time [5].

It is in this context that we present a proposal of an integrated front-end architecture for in vivo customized detection. A new and challenging scenario defined as the pervasive system is focused on the development of systems capable of monitoring human bodily functions and to transmit the resultant data for a clinical patient's monitoring [6]. Thanks to this approach, it could be possible to monitor patients anywhere and at all times with important impact on their quality of healthcare preventing the worst scenarios for the patients as well as improving the wellbeing and continuing activity of the whole population. The possibility of controlling how a therapy is working, detecting symptoms, and knowing how the disease is progressing will improve the personalized medical care known as theranostics. Patients at risk because of their genetic background, chronically ill or elderly people will be monitored outside of and beyond visits to the hospital or at the surgery. Here, the significant advantage is to monitor patients in their routine daily activities, as traditional clinical monitoring would be replaced by continuous and remote monitoring [7]. This could have a great impact on patients' quality of life and could reduce the cost of the overall healthcare system [8]. Across all medical applications and diseases, findings suggest that chronic illness deserves special attention [9], particularly in the case of cardiovascular illness [10].

With the aim of medically monitoring the patients, there are two different approaches that are typically used: external body sensors, and implantable devices, i. e. , non-invasive approaches versus the invasive ones. In the case of external sensors for non-invasive physiological monitoring [11], a multiplatform is suggested [12], with particular interest in the wearable solutions and unobtrusive sensing methods [13,14] and, in particular, the most recent advances in textile-based electronics are relevant [15]. In the case of the invasive techniques that are the focus of this document, the type of solutions that have been developed, and those which are currently in progress [16], have as a classic example the cardiac implant, which was the initial application of these devices, now with advanced capabilities such as recently reported by Lee et al. [17]. The evolution of semiconductor technology, with low-voltage and low-power electronics, allows the integration of several implantable devices for different functions. These approaches could also be combined in order to define a body sensor network (BSNs) [18]. The placing of a central control node that acts as a master node, with other slave nodes located on or inside the body, monitoring different vital signals, defines a typical wireless network that could fulfil the theranostics needs of the patients.

Theranostics covers a wide range of applications as health interventions with drugs (pharmacogenomics), nutrition (nutrigenomics) and vaccines (vaccinomics), as well as diagnostics for human diseases [19]. Implantable medical devices are widely used for therapeutic [5] or life-saving purposes such as cardiac arrhythmia, diabetes, and Parkinson's disease [20]. Applications include drug delivery systems, pacemakers, implantable cardiac defibrillators (ICDs) and Neurostimulators [1]. Some real-time monitoring applications include physiological parameters like blood pressure, glucose levels and collecting data for further analysis [5].

These devices often contain electronic components that perform increasingly sophisticated sensing, computation, and actuation, in many cases without any patient interaction [1] as in the applications mentioned above, performing complex analyses with sophisticated decision-making capabilities. They are capable of storing detailed personal medical information, and communicate automatically, remotely, and wirelessly [2]. Implanted biosensors form a wireless network that can be used for data aggregation and data dissemination applications [5].

The system introduced in this paper is conceived to be implanted under the human skin. The powering and communication between this device and an external primary transmitter are based on an inductive link [21]. The design presents two different approaches: defining a true/false alarm system based on either amperometrics or impedance into a grid of nano-biosensors that could permit the monitoring of several diseases by in vivo analysis of the corresponding biomarkers.

2. Description and Challenges of a Customized Biomedical Implantable Device.

2.1. State-of-the-Art of the Multipurpose Diagnosis Implantable Devices.

Many different problems need to be overcome in obtaining the ideal implantable device [22]. First and foremost, the device must be biocompatible to avoid unfavourable reactions within the body. Secondly, the medical device must provide long-term stability, selectivity, calibration, miniaturization and repetition, as well as power in a downscaled and portable device. In terms of the sensors, label-free electrical biosensors are ideal candidates because of their low cost, low power and ease of miniaturization. Recent developments in nanobiosensors provide suitable technological solutions in the field of glucose monitoring [23,24], pregnancy and DNA testing [25], and microRNA detection [26]. Electrical measurement, when the target analyte is captured by the probe, can exploit voltometric, amperometric or impedance techniques. Ideally, the device should be able to detect not just one target agent or pathology, but rather several different agents and it should be capable of working in a closed-loop feedback, as described by Wang [27] in the case of glucose monitoring.

Several biomedical devices for in vivo monitoring are currently being developed [24,28,29]. Thus, highly stable, accurate intramuscular implantable biosensors for the simultaneous continuous monitoring of tissue lactate and glucose have recently been produced, including a complete electrochemical cell-on-a-chip. Moreover, with the parallel development of the on-chip potentiostat and signal processing, substantial progress has been made towards a wireless implantable glucose/lactate sensing biochip [30]. Elsewhere, implantable bio-micro-electromechanical systems (bio-MEMS) for the in situ monitoring of blood flow have been designed [31]. Here, the aim was to develop a smart wireless sensing unit for non-invasive early stenosis detection in heart bypass grafts. Interestingly, this study examines the use of surface coatings in relation to biocompatibility and the non-adhesion of blood platelets and other blood constituents. In this case, the nanotechnology as a KET seems to be a useful tool for improving the biocompatibility of silicon bio-MEMS structures.

A theranostic device has one or more specific molecular recognition markers for cells on the surface thereof, wherein the recognition markers are selected from the group consisting of peptides, proteins, antibodies, antigens, aptamers, molecular imprinted polymers and polynucleotides. When the device is implanted in a body, cellular ingrowth is controlled, with desired cell types anchoring and proliferating on the implant's surface to generate a thin layer, and thereafter ceasing accumulation. The cellular layer thereby presents a biomimetic surface acceptable to the body, and also presents a low barrier to diffusion of analytes with at least substantially constant diffusion characteristics, allowing the use of an analyte sensor within the article.

In biomedical research, there is a great need for multipurpose, reliable, and possibly implantable telemetric tools. By using sensor inputs, such devices allow the automated gathering of information on physiological parameters without restraining or stressing their subjects. For this purpose, a versatile implantable and four-channel telemetry data-acquisition unit was implemented by Wouters et al. [32], in a 2-pm n-well CMOS process. The dimensions of this single-chip implementation are 4.7 × 7.1 mm 2 . In the form of an implantable or portable telemetry system, a low-power mixed analog-digital CMOS integrated circuit combining several sensor interfaces, the processing and control circuitry, and the telemetry unit is intended for monitoring body temperature and physical activity.

A versatile theranostic system was developed by Young Choi et al. [33] for the early detection, targeted therapy, and therapeutic monitoring of colon cancer, by using poly(ethylene glycol)-conjugated hyaluronic acid nanoparticles (P-HA-NPs) which can selectively accumulate in tumor tissue. For the diagnostic application, a near-infrared fluorescence (NIRF) imaging dye (Cy 5.5) was chemically conjugated onto the HA backbone of P-HA-NPs. Arjang Hassibi has worked in the areas of biosensors and bioelectronics, biomedical electronics, and integrated circuit design. His company, InSilixa, is working on a multi-diagnostic system, using semiconductor-integrated DNA-sequencing technologies to create point-of-care diagnostics devices. The idea is to take advantage of “large-volume semiconductor technology,” manufacturing systems that are widely available and well established, to gain economies of scale [34].

An RF-powered wireless three-channel implantable bio-sensing microsystem has been developed with blood pressure, EKG, and core body temperature sensing capability for untethered genetic tests. A flat silicone blood pressure sensing cuff with a MEMS capacitive pressure sensor is employed to form a novel less-invasive blood pressure sensor, which avoids vessel occlusion, bleeding, and blood clotting associated with the conventional catheter-based sensors. The implantable microsystem can be powered by an adaptively controlled external RF energy source at 4 MHz to ensure a stable on-chip power supply. On-going research efforts are devoted to demonstrating in vivo performance in laboratory animals [35].

Finally, a patent (US 8750961 B1) with a multi-axis, multi-purpose sensor for use with implantable medical devices, and for simultaneously detecting the patient's posture and activity level has been developed [36]. The sensor includes a hermetically sealed, fluid-tight, bio-compatible housing. The housing is formed of a plurality of adjacently secured sides, and a plurality of side electrodes coupled to the sides. A central electrode is disposed at the geometric center of symmetry of the housing, to allow measurement of voltage changes between the central electrode and the side electrodes. A non-toxic electrically conductive electrolyte fills about half the housing, and immerses part of the central electrode and the side electrodes. The sensor further includes a low frequency bandpass filter for passing low frequency signals indicative of the patient's posture, and a high frequency bandpass filter for passing high frequency signals indicative of the patient's activity.

2.2. Research Challenges in Implantable Devices.

The new generation of implantable devices must overcome some main barriers at its conception stage as for example: size, energy available, power dissipation, power management, signal processing, communication of the measured data, bio-compatibility, chip-level integration, packaging, bioethics, and biosecurity [37]. A conceptual body map of commercial and in development phase implantable sensors is depicted in Figure 1 , focussed on the most relevant disease processes based on Oesterle et al. [16].

There are very interesting implementations that combine ASICs, MEMS, and the design of integrated antennas for the RF powering of the system and the telemetry of the data, based on a link between external and internal coils. A fully wireless implantable cardiovascular pressure sensing system was developed by Chow et al. [38], combining a 130-nm technology and a MEMS capacitive sensor, powering the system through an external 35-dB-m RF powering at 3.7-GHz, with a distance range up to 10 cm inside the body, with a telemetry capability of 42.2 kb/s of channel data-rate, operating at 2.4-GHz and with a medical stent of 3 cm long. Cleven et al. [39] have also developed another interesting application regarding cardiovascular problems, where an implantable wireless system for monitoring hypertension is presented. The capacitive sensor, which is based on a MEMS implementation in a metal capsule, and the electronics (ASIC), forms a tip of 20 cm that will be placed in the femoral artery. In this case, the telemetry and powering is fixed at 133-kHz, with a maximum distance of 10 cm.

In Majerus et al. [40], a bladder-pressure-sensing implantable for chronic patients is introduced, based on a specific ASIC design, and also based on a RF powering (LC coupling at 3-MHz), and telemetry solution, but operating in an unlicensed ISM band (27.12-MHz), and a rechargeable battery solution. In this case, the size is also based on medical constraints given by minimally invasive cystoscopic surgery, defining a final capsule of 7-mm wide by 4-mm thick by 15-mm long. Another interesting implementation is related to glaucoma [41]. In this case, an RF intraocular pressure monitor is implemented based also in a MEMS solution for the sensor and in RF wireless transmission at 2.4-GHz and RF powering operating at 3.65-GHZ. In this way, RF powering and telemetry path are isolated in the same way as in the other examples. An additional approach is presented in Pivonka et al. [42], where RF powering and biomedical telemetry at 1.8-GHz are combined with the aim of developing a locomotive implantable.

Other interesting examples for close-loop systems are presented in [17,43,44]. Salam et al. [43] developed an implantable drug delivery system for the treatment of refractory epilepsy. The system is able to acquire real-time epileptic detection with focal antiepileptic drug injection feedback, combining electronics, pumps, reservoirs, etc. Lee et al. [17] presented an implantable microstimulator applied in a closed-loop cardiac pacemaker. Monge et al. [44] reported a fully intraocular epiretinal prosthesis, based on a 65 nm CMOS ASIC with 512 independent channels, integrated with a flexible MEMS origami coils, for the inductive powering at 10-MHz, and telemetry at 160-MHz, and parylene substrate to provide the intraocular implant.

Such complex systems are developed combining different Key Enabling Technologies (KETs) which are main contributors to overcoming the challenges involved in the development of an implantable device. Figure 2 introduces the suggested share of cross-cutting KETs involved in the development of a nano-enabled implantable device for in vivo biomarkers monitoring.

Current research is focused on the miniaturization and progression of the implants [45], in pursuit of less powering and long monitoring devices. The ongoing evolution in this field is based on the higher technological capabilities of microelectronic technologies, with higher density of integration, and involving a blend of MEMS, packaging and interconnects. The possibility of integrating new dedicated miniature transducers, such as pressure sensors [46], for arterial blood oxygen saturation, and accelerometers [47] for heart monitoring or cochlear implementations [48] are present examples. The progress in miniaturization of lab-on-chip solutions and integrated optics [49] opens the possibility of advances in new implantable medical devices and new challenges, culminating in the theranostic approach where the implantable will be able to deliver drugs. The system will implement the right algorithms for the control of drug delivery as well as the suitable reservoirs and pumps. This approach is feasible with the evolution and progress of MEMS, not particularly from silicon but from flexible polymers, and in terms of the lab-on-chip solutions in the field defined by medicine as micromachines [50], where there is an active control of fluids. The objective is to deliver drugs in a better way, more focalized in the local area or target of interest, rather than through traditional oral medication. The system should close the loop, with a monitoring and actuation role. The great paradigm is the artificial pancreas, with the design of an implantable system to monitor the glucose level, and a pump injector system. The combination of these enabling technologies creates the possibility of moving forward with advanced solutions such as artificial organs [50].

In the case of the optic approach, there is a significant limitation of integration, in terms of space, and also in terms of its implementation cost, but it is an interesting field of development. This approach has some important advantages compared with sensors based on an electrical measurable parameter, current or voltage, that make it interesting for integration in future developments. Among the benefits are its immunity to electromagnetic radiation, temperature tolerances, fewer risks to biological tissues and reliability when working in aqueous solutions. An interesting example is presented by Bingger et al. [46], where MEMS and an optoelectronic solution are implemented for continuous long-term monitoring of vital medical parameters such as arterial blood oxygen saturation, pulse and respiratory frequencies.

2.2.1. The Powering Module.

A main issue is the way the implant is powered. There are different options, always determined by the location of the packaged system, and the available area or volume location for the implantable. The first option is to feed the system through a battery [37]. In the field of battery-oriented biomedical and implantable devices [51], the classical approach is based on lithium batteries size C or D, summarized in Table 1 .

Current implementations of communication links in implantable devices are not suitable for many applications because of their poor harvesting efficiency [52]. Energy harvesting solutions must explore if a battery is not an affordable solution: these are defined as self-powered solutions [53]. The first solution could be based on electromagnetic induction [54], with various approaches. Some solutions propose the implementation of coils on a PCB substrate [55], or coaxial aligned coils with or without a ferrite rod [56], with all the attendant problems related to misalignments between the primary powering coil and the implanted secondary coil [57] and electronic implementations for a dynamic control of the power and voltage generated in the implantable in terms of the actual magnetic field that is generated [58]. On the one hand, we have an external element that plays the role of energy wireless power source, based on a class E amplifier, powered by the external battery, which supplies power to the implantable device through the skin [59,60] (see Figure 3 ). This powering is local, working for instance in the 13.56 MHz ISM (Industrial, Scientific, Medical) band [55], powering 10 mW at a distance of 10 mm between coils. The implantable operating at this ISM band must be placed near to the external generator. In Kilinc et al. [61], a particular case is presented. A wireless power-transfer in vivo implantable device for free moving small animals is derived. The scenario is that the living space for the animal is transformed into a full powering base.

However, this is one approach for the ISM band. The 13.56 MHz is a very low value. The more usual bands, taking into account the need to reduce the size of the implanted antennas and locations to place the implants in the body [62], are the bands of 433 MHz, which have similar results to the 402‘405 MHZ MICS (Medical Implant Communication service) band [63], 915 MHZ, 2.45-GHz and 5.8-GHz. Currently, there are more works examining the optimization of the design of the implanted antennas [64] and the need to analyse the transmission losses between the external antenna and the deeply implanted antenna [65].

A particular example of this is introduced in Zhang et al. [66]. There, the rectifier module is designed to work at the 915-MHz ISM bandwidth (only in region 2). The performance of the RF source is quite small, from a typical 4 μW/cm 2 for GSM to 1 μW/cm 2 for the WiFi band. For coils, typical values are lower than 1 μW/cm 2 , but as much as 1 mW for close inductive coils (a few cm). In the 915-MHz ISM bandwidth, at 1.1 m, the energy recovered is around 20 μW [67]. New approaches are being developed. In particular, the use of ultrasonic powering instead of RF powering is of great interest. In Zhu et al. [68] and Moheimani et al. [69], 1 V is generated with a power capability of 21.4 nW.

Nevertheless, there are other approaches to power an implantable device without the use of a battery. One approach is based on the vibration energy harvesting point of view and the use of MEMS [53]. In Abidin et al. [70], a MEMS piezoelectric generator is used to harvest energy from vibrations; it also uses supercapacitors as storage elements. An example of a MEMS designed for implantable devices is given in Martinez-Quijada and Chowdhury [71], where it is stated that the micro-generator is able to generate more energy per unit volume than conventional batteries; that is, an RMS power of 390 μW for 1 mm 2 of footprint area and a thickness of 500 μm, which is smaller than the volume of a typical battery in a pacemaker. Another approach is based on the use of fuel cells. The conception of a fuel cell as a biogenerator for implantable devices has emerged, with interesting results like in Zebda et al. [72], where a primary glucose fuel cell is derived for an implantable device. In some ways, the basic concept is the use of fluids in the body as a fuel source for the fuel cell, which would be an inexhaustible energy source. An interesting approach is the use of glucose as a fuel source, or the oxygen dissolved in blood [73,74]. Advanced approaches also explore a shift to the use of white blood cell capacities in biofuel cells [75]; or approaches such as that in Siu and Chiao [76], where the fuel cell is based on the use of a microorganism to convert the chemical energy of glucose into electrical energy, in a PDMS structure.

2.2.2. The Encapsulation of the System.

Bio-compatibility of the final device is a main barrier and challenge for the spread of implantable sensors. The encapsulation has to satisfy different properties, especially with regard to its lifetime. For instance, it has to be biocompatible and have a low dielectric constant [77], as well as being conformal and resistant.

Implantation of synthetic medical devices generates an immediate and complex material-related inflammatory response, such as blood and tissue incompatibility and bio-fouling [78]. Biofouling of the sensor membrane is an important cause of sensor dysfunction [79]. Therefore, the design of implantable BioMEMS devices must reduce this immune impact, minimize bio-fouling, reduce the physical effect of the implant on the surrounding tissues and reduce the degree of cell adhesion achieved by the implanted device. To avoid these adverse physiological effects, the implanted devices must be packaged with bio-compatible materials. However, bio-compatible materials might not always be compatible with the device requirements [78].

Currently, common and widely used materials in implanted biomedical devices with high compatibility are polyethylene glycol (PEG), polydimethylsiloxane (PDMS), PTMO (poly tetramethylene oxide [78] and parylene-C [80]. Polymer coatings are used for glucose sensors as they reduce the diffusion of interferences to the sensor while simultaneously balancing glucose and oxygen diffusion to enable an adequate glucose response. They are durable, inert, and capable of tolerating harsh environments produced by the FBR. Commonly evaluated polymers are, Nafion, polyurethane, polyethylene glycol (PEG), and hydrogels. Nafion is a perfluorosulfonic acid-based polymer that has been implemented as a bio-compatible coating. Polyurethane (PU) has been used extensively as an outer membrane to act as a bio-compatible interface with the surrounding host tissue. Surface passivation with polyethylene glycol (PEG) has been a widely studied strategy for resisting bio-fouling [81]. Hydrogels have a modulus similar to subcutaneous tissue and absorb water readily allowing easy diffusion of analytes to a sensor.

In vitro analysis in an osmotic glucose sensor evaluated identified 15 potential candidate materials which are shown in Table 2 below [79].

Other biocompatible materials include collagen layer for encapsulation [81], or gold, silicon nitride, silicon dioxide and SU-8 for coating use, able to reduce biofouling [82]. Coating of silicon carbide for example, can be used to significantly reduce thrombus formation on the surface of the devices, especially if the device is exposed to blood [83]. Bouaidat et al. [84] have also mentioned the use of phosphorous glass (SiPOC) for cell adhesion in BioMEMS.

The application of NDGA-crosslinked collagen scaffolds is also a good method for enhancing the function and lifetime of implantable bio-sensors by minimizing the in vivo foreign body response. Ju et al. [85] have developed a 3D porous and bio-stable collagen scaffold for implantable glucose sensors. The scaffolds were fabricated around the sensors and crosslinked using nordihydroguaiaretic acid (NDGA) or glutaraldehyde (GA) to enhance physical and biological stability. كيم وآخرون. [86] reported an implantable sensor for real-time monitoring of the changes in bladder volume with PDMS and parylene-C. They find that both can be used as safe coating materials for the implantable bladder volume sensor reported.

A novel polymer coating consisting of poly(lactic-co-glycolic) acid (PLGA) microsphere dispersed in poly(vinyl alcohol) (PVA) hydrogels was evaluated in combination with dummy sensors as a “smart” drug eluting bio-compatible coating for implantable biosensors to prevent the foreign body response, and thus enhance sensor performance in vivo [80]. Single or multiple electro-spun layers can be used to address mass-transport limiting and additional membranes for improving biocompatibility of implantable biosensors and other biomedical devices requiring analyte transport, especially the first generation implantable glucose biosensors [87].

In summary, packaging techniques used must assure a long-term stability and surgical risks must be avoided. To fulfil these requirements, available implants in the market typically use hermetic packaging in laser-welded enclosures [88]. Nevertheless, for the envisaged miniaturized implants, where cans and micro-lids are used, this solution takes too much space. In that case, implantable devices for sensing and therapeutic purposes with active regions fully exposed to the physiological environment are a great challenge [89]. New approaches based on thin-film coating solutions are in progress to overcome these problems [90,91]. In Xie et al. [90] a bilayer solution based on an atomic layer deposited (ALD) Al 2 O 3 combined with Parylene C for long-term encapsulation is presented, and in Sutanto et al. [91], a packaging and non-hermetic encapsulation MEMS flip chip technology for implantable devices is developed.

2.2.3. The Nano-Biosensor.

Special attention must be focused on nanobiosensors [92]: they need to combine accuracy, reliability, precision, life span, manufacturing and scalability, as well as address wealth and environmental risks, in order to overcome technological and market bottlenecks. A nanobiosensor or nanosensor is generally defined as a nanometre size scale measurement system comprising a probe with a sensitive biological recognition element, or bio-receptor, a physicochemical detector component, and a transducer in between. Two types of nanosensors with potential medical applications are cantilever array sensors and nanotube/nanowire sensors and nanobiosensors, which can be used to test nanolitres or less of blood for a wide range of biomarkers.

Then, a biosensor is a measurement system for the detection of an analyte that combines a biological component with a physicochemical detector. The general function of a biosensor is to convert binding events between biological receptors and target agents into a signal thanks to a transducer which can be based on an optical, a thermal, a gravimetric or an electrochemical detection (see Figure 4 ). This last category has gained increasing attention in the last few years. The high sensitivity, low cost and easy miniaturization of the electronic detection taken in conjunction with the wide range of applications, has resulted in these devices becoming a perfect analytical tool in different fields, such as diagnosis of genetic diseases, detection of infectious agents, study of genetic predisposition, development of personalized medicine, detection of differential genetic expression, drug screening, etc.

The development of highly sensitive and low-cost sensors in the nanoscale, and its combination with nano-microfluidics solutions [94], based on micro-channels, micromixers and microvalves, are increasing the interest in the implementation of multi-parametric point-of-care devices, as a portable and low-cost solution to enhance diagnostics methods. In summary, Figure 5 shows principal technologies, challenges and materials for multipurpose implantable sensors.

3. Conception of the Bio-Implantable Customized Multi-Sensor.

3.1. A Multipurpose Biosensor Architecture.

Instead of defining a particular architecture of the implantable device for each sensor, the new approach in this paper introduces the design and use of a general architecture that will require minor modifications for a final customized implantable device which could be suitable for a set of specific applications.

The objective is to have a generic array of nanosensors (electrodes) as an implantable system. Figure 6 shows the combination of cell clinic solutions concept as a lab-on-a-chip and electrical sensing techniques in a single implantable device.

The envisaged concept is applied in the definition of an on-chip configurable array of biosensors. This configuration will take place before the implantation thanks to a standard programmable bio-nano-chip approach [34]. A modular standard lab-on-a-chip approach will be followed to adapt the sensors in a quick, efficient and reliable way and then the implantable system will be placed into the patient ( Figure 7 ). This concept of programmable platform could be adopted with the aim of developing a POC external device. These electrodes will be functionalized before the implantation in the human being thanks to the microfluidics (inflow/outflow) circuitry. Afterwards, the sensors will be checked and the chip cleaned and ready for the implantation.

The system will be enabled thanks to a system-on-a-chip (SoC) technology. CMOS microelectronics, MEMS and microfluidics will be combined to implement the programmed implantable, and easily adapted for the specific needs of the patient. The generic ASIC will combine the integrated electronics with an array of nano-biosensors (Sensors array), depicted as electrodes in Figure 6 , which would be functionalized for particular purposes [98]. Generic modules for the power management, narcoleptic system design (NSD), communications, signal processing, the processor and data logging will be integrated to fulfil time-to-market constraints.

Tsai et al. [99] addressed the concept of the envisaged integrated multi-analyte biochip for an implantable device, in terms of the fabrication, where microfluidics (PDMS micro-channel), and a dielectrophoresis concentrator (DEP) are combined with external discrete electronics. The aim of the microfluidic system is to prepare and transport the fluid into the microcapillaries. Then, the preparation step consists in the separation of the fluidic and/or suspended particles [100], the mixing of the fluids for cell activation and mixing reactants for initiation. It could take place along the capillaries or inside of created droplets. These droplets are also useful to encapsulate biological particles or chemical reagents. In some cases, the sample also needs to be focalized [101] before it flows through the electrical or optical detection system as seen in Figure 7 .

Based on the concept of Tsai et al. [99], the implantable multi-purpose sensor will be defined by the combination of configurable sensors as, for instance, glucose sensor [102,103], thermal metabolic sensor [104], PH and other sensors to detect the concentration of molecules, typically metabolites, such as glucose, lactose, sodium or ATP as examples of endogenous molecules, or exogenous molecules, such as etoposide and ifosfamide.

3.2. The Electronic Design.

The envisaged integrated electronics is depicted in Figure 8 . The ASIC will combine all the necessary electronic modules with the sensors' array of the functionalized biosensors. When the implantable is placed in the body, a programmed check of the state of the biosensors should take place. The system will check the sensors' array during the implantable life, and send a critical message to the final user if a malfunction is detected and the implantable must be removed while it is implanted.

The system will be based on the use of two different antennae, but it could be based on just one: one will be working at a lower frequency to harvest energy (power link), based on the previously presented concept of inductive powering, and a second antenna operating at higher frequencies for the communications (communications link). In this case, the communication link can be established around hundreds of MHz (usually in the 400 MHz ISM band) allowing higher communication rates and reducing the size of the antenna, as previously stated. The first antenna is focused to power the electronics through a dedicated inductive link operating at lower frequencies than the communication antenna. In that way, each antenna can be optimized for its functionality.

It is also possible to use just an inductive link for both purposes, and bi-directionally transmit the data [57]. However, the amount of transmitted information is limited and the size of the antenna is considerably bigger. The communication set-up could be based on a simple backscattering, defining an AM modulation, which is the approach taken. In our first ASIC implementation an inductive link for both purposes, operating at 13.56 MHz, was implemented. This is a good value for low power emission and appropriate to a subcutaneous placement. In our design a planar rectangular coil of 5.5 mm × 14.5 mm with a thickness of 0.5 mm has been designed, as a proof of concept for the antenna. It has seven turns with a conductor width of 0.2 mm. The design presents an inductance of 400 nH and a series resistance of 340 mΩ.

An AC/DC integrated rectifier generates an unregulated DC voltage from the electromagnetic energy delivered through the inductor link in the Power Management Module. The AC/DC block is based on a half-bridge NMOS rectified with a bulk control voltage.

The system has a power-on-reset module (POR) that activates the electronics when enough energy has been recovered through the inductive coupling. A LDO and a low-voltage low-power band gap reference circuit generate a DC regulated voltage to drive all the on-chip electronics. A NSD module is also implemented to enable the different modules thanks to the POR and the BG. The combination of these modules defines the Power Management Module.

Afterwards, the integrated electronics is introduced to drive the biosensor, make the measurement and to generate the data to be transmitted (Sensors Signal Conditioning). Usually, a low-voltage, low-power potentiostat circuit or similar instrumentations are used to control each sensor of the array (Sensor Control Potentiostat). CMOS electronics will be implemented for each of the sensor's array (Chanel Sensor) [105‘107], combining different sensing techniques, such as chronoampetometry (CA) and cyclic voltammetry (CV), for sensors' characterization and calibration tasks, or electro-chemical impedance spectroscopy (EIS) generated by the Signal Generation Module, which will have the capability to generate DC voltages, a DC sweep or AC signal in order to cover the different techniques. In this case, we focus our attention on DC internal voltages which are designed to fix a DC voltage for the sensor. Three internal voltages of 0.6 V, −0.6 V, and 0.5 V can be selected. These signals are generated from the regulation module, based on the implemented band gap reference circuit. In our case, for a three electrode case, a low-voltage low-power CMOS potentiostat amplifier was implemented for an amperometric measurement.

These voltage levels could be applied by the potentiostat amplifier to the three electrode biosensor, defined by: (a) the working electrode (W), which serves as a surface on which the electrochemical reaction takes place and will be functionalized by the lab-on-a-chip module depicted in Figure 6 and Figure 7 ; (b) the reference electrode (R), which measures the potential at the W electrode; and (c) the auxiliary or counter electrode (A/C), which supplies the current required for the electrochemical reaction at the W electrode. A single potentiostat amplifier occupies an area of 327 μm × 260 μm, and has an average power consumption of 51.2 μW, which is smaller than Paglinawan [108] which has an area of 0.16 mm 2 , and a power dissipation of 600 μW, or Ahmadi and Jullien [109] which has a power dissipation greater than 150 μW. Its open-loop gain is 60 dB at low frequencies, and 50 dB1 kHz.

The current that is generated in the amperometric sensor, which is proportional to the electrochemical reaction that is generated at the working electrode, is measured by a transimpedance amplifier (TIA). Its input resistance of the design is 1 GΩ@DC, allowing a current detection up to 1 nA. The current-to-voltage conversion is defined as V TIA = −I W R TRANS , where I W is the current through the working electrode and R TRANS is the externally selected gain resistance. A second gain stage based on an inverter configuration follows the TIA and adapts the voltage values for the next stage, defining the Sensor Conditioning module.

The measured signal is forwarded to the Data and Modulation Processing modules. In this case a simple absence/presence detector is defined in the Chanel Sensor Module. The detection is based on the conception of an event-detector and the True/False detector works as an alarm: when the analyzed concentration level exceeds, under or over, a threshold value or the system detects the alarm condition, then the modulation process is activated to send the information to the external reader using a backscattering method through the inductive link, which can be AC or DC modulation.

3.3. النتائج.

A bipolar power scheme able to supply a regulated differential voltage of ±1.2 V and a maximum current of ±1.5 mA has been implemented. The Texas Instrument ® TRF7960 is used as external reader with a maximum emission power of 200 mW at 13.56 MHz. The desired on-chip regulated voltages of ±1.2 V are obtained for a distance up to 20 mm on air between coupling antennae. This analysis has been carried out in terms of the distance (Z-axis), between the external antenna and the coil designed in the PCB which defines the full implantable, that is, between the reader and the implantable. However, it is also necessary to have an approach to the misalignments between both antennas in the XY plane. Figure 9 depicts the rectified voltage (V rec ) distribution in function of the XY misalignment for Z distances of 10, 15, and 20 mm. It can be noticed that the further the antenna is placed from the centre the lower the rectified voltage is. A more accurate study with human tissue is beyond the scope of this study.

A suitable solution for the detection of threshold values is based on the use of comparators, in terms of silicon area and power consumption, to detect one or several threshold values with medical interest. In the implemented approach, some comparators capable of detecting three different threshold voltages (V th1 , V th2 and V th3 ), generated on-chip, have been implemented. These values are used to define a simple AM modulation protocol.

The signal is always a high level “1” but when a threshold value is achieved then a “0” level is generated. This functionality is based on the use of the comparators, monostables flip-flops and a very simple digital circuitry. As soon as there is enough voltage, the Power-On-Reset module generates a signal that activates the circuitry and the antenna starts to transmit continuously a series of “1”. When the first threshold level is achieved, the system transmits one zero (T th1 ). If the second is reached, two zeros are transmitted (T th2 ), and when the third is achieved a series of three zeros are sent (T th3 ). A zero time slot interval is defined as 250 ms (T th1 = 250 ms). In this way, the external reader can be quickly advised every time the desired substance exceeds the programmed threshold level or levels.

The instrumentation and the communication protocol were validated using several concentrations of K 4 [Fe(CN)] 6 in PBS. In this case, a commercial sensor was used [60]. Several cyclic voltammetries (CVs) were carried out in order to verify the performance of the Control and conditioning modules. These measurements were compared with those obtained with a commercial potentiostat amplifier, the CH 1232A from CHInstruments®. These measurements were also used in order to calibrate the setup, and check the obtained values of the measured current peaks for the oxidation peak (around 240 mV), and the reduction peak (around 170 mV), for each concentration of K 4 [Fe(CN)] 6 in PBS tested, from 1‘5 mM. Also, this setup was used to validate the measured CV shapes obtained by the commercial equipment and the full-custom implementation. After the CV characterization some amperometric tests were done for different concentrations of K 4 [Fe(CN)] 6 in PBS: 1 mM, 2 mM, 3 mM, 4 mM and finally 5 mM. Then, an experiment was carried out where the concentration was changed from 1 to 5 mM in time, with a fixed voltage of 500 mV in the sensor. This voltage is defined not at the oxidation peak. For this value of voltage applied in the sensor, the current varies from an average current of 3 μA (1 mM), to 16 μA (3 mM), up to 28 μA (5 mM). This experience was then carried out to validate the detection protocol, for a particular case implemented based on three threshold values. These values were programmed to detect the variations in the concentrations, defined by: Vth1, Vth2, Vth3, as is depicted in Figure 10 , taking into account the current expected for each concentration case and defining different windows of comparison. When the first threshold is detected, then a first zero is transmitted, with a programmed width of 250 ms. When the second threshold level is reached, then two zeros are transmitted, in this case with an amplitude of 500 ms. Finally, in the particular case that a threshold Vth3 is defined to detect the highest concentration level, the modulation and data processing module will generate the longest transmission of zeros, in this case, three zeros with a total width of 750 ms.

4. Market Approach and Discussion.

4.1. Innovation and Commercialization Chances in a Multi-KETs Scenario.

In September 2009, the European Commission published its communication “ Preparing for our future: Developing a common strategy for key enabling technologies in the EU ” [110]. This strategy identifies the need for the EU to facilitate the industrial deployment of KETs in order to make its industries more innovative and globally competitive. KETs are one of the key factors in realizing the overall policy objectives of Europe 2020, due to the importance of these technologies for the competitiveness and innovation of European enterprises as well as for the development of sustainable products and processes [111]. In this context, Horizon 2020, the biggest Framework for Research and Innovation, has scheduled over 74 billion € for research funding focused on three fundamental pillars: 24.598 million € intended for Scientific Excellence, 31.748 million € for Society Challenges and 17.938 million € for Industrial Leadership. The last one aims to support SMEs in the industrial development and application of KETs , considered crucial accelerators for innovation and competitiveness [112].

KETs have been selected according to economic criteria, capital intensity, technology intensity, and their value adding enabling role [113]. The six KETs are: Nanotechnology, Micro and Nano Electronics, Photonics, Advanced Materials, Biotechnology Industry and Advanced Manufacturing Systems [114]. Among them, Nanotechnology is one of the most promising KETs due to its economic and social growth potential, since it has been considered the greatest impulse to technological and industrial development in the 21st century and the resource for the next industrial revolution [115‘118].

The integration of different Key Enabling Technologies (KETs) represents a vital activity in H2020. About one third of the budget assigned to KETs will go to supporting innovation projects integrating different KETs [119]. Cross-cutting KETs activities will in general include activities closer to market and applications. The global market volume in KETS are 646 billion euros and substantial growth expected is approx. 8% of EU GDP by 2018 [113]. In the Healthcare domain, short (2017) and medium (2020) perspectives of cross cutting KETs are shown in the Figure 11 .

The European Commission stated that the EU has very good research and development capacities in some key enabling technology areas, but it has not been as successful in translating these results into commercialized manufactured goods and services [110]. R&D projects implemented in FP6 and FP7 frameworks have successfully delivered a lot of new nanomedicines but few products onto the market. In this context, the Commission states that bridging the so called “Valley of Death” to upscale new KET technology based prototypes to commercial manufacturing, often constitutes a weak link in the successful use of KETs potential. This is meant to be the “European Industrial Renaissance” by covering the whole value chain from Lab-to-Market as the principal aim of H2020 [113].

4.2. توقعات السوق.

The emerging sector of applied nanotechnology is addressed to biomedicine (nanobiotechnology and nanomedicine) which is the area of greatest projection of the future [120]. There are currently 247 nanomedical products that have been approved or that are in several stages of clinical trials. Industry market reports describing companies and their products related to nanomedicine and nanobiotechnology have also increased in the last several years [121]. It is expected that the annual global market for nano-related goods and services will top $3 trillion in 2020 [122]. Beyond, the medical sensors global market is expected to reach 15.5 USD billion in 2019, growing at a Compound Annual Growth rate (CAGR) of 6.3% from 2018 to 2019 [123]. Findings suggest that market growth for biosensors and biochips is virtually exploding. There are markets for biosensing technologies in the Asia-Pacific region, which show Compound Annual Growth Rates of 11% (2008‘2018). Growth Rates of 10.7% occur in the highly developed market of the United States (US). In fact, this market is projected to reach $8.5 billion in US currency within five years, in about 2018 [124]. On the other hand, the global market for theranostic nanomaterial was valued at $112 billion in 2018 and is expected to reach $188 billion by 2017, registering a five-year CAGR of 10.8% for the period 2018‘2017 [125].

Today, the implantable medical device market is oriented to the increasing elderly population and the associated increase in the prevalence of chronic degenerative diseases. However, the use of microtechnologies and MEMS in implantable devices is still in its infancy with few technologies currently approved for marketing in the US [126]. There is no identifiable market in the private sector for personalized and precision medicine yet [127]. The Food and Drug Administration (FDA) regulatory process will determine the concrete translation from benchtop research to commercialization of implantable nanosensors through clear and reasonable regulations. In this context, the FDA is collaborating with the interagency National Nanotechnology Initiative (NNI) to help formulate its guidelines with respect to many aspects of nanotechnology in the realms of cosmetics, diagnostics, and therapeutics [128].

4.3. Ethics Concerns.

Designers of implantable medical devices have balanced safety, complexity, power consumption, and cost. However, today there are new concerns to take into consideration: security and data privacy [1]. As biosensors monitoring involves collection of data about vital body parameters from different parts of the body and making decisions based on it, the information is of a personal nature and is required to be secure [5] . The reason is to protect patients from acts of theft or malice, especially as medical technology becomes increasingly connected with other systems via wireless communications or the Internet. Implantable medical devices, including pacemakers, cardiac defibrillators, insulin pumps, and neurostimulators feature wireless communication [129].

Susceptibility to security breaches could compromise performance safety and the privacy of patients [2]. Burleson et al. [1] stated that there are two types of vulnerability: privacy, in which patient data is exposed to an unauthorized party, and control, in which an unauthorized person gains control of the device's operation or even disables its therapeutic services.

There is a need to ensure the privacy and security of medical data [4]. Recent analyses of implantable medical devices have revealed several security and privacy vulnerabilities [1]. For example, wireless connectivity could compromise the confidentiality of transmitted data or send unauthorized commands to the device [129]. Privacy specifications seem to be vague [4], in fact medical devices vary widely with regard to security features because no specific security guidance or requirements have been promulgated by the FDA [2].

Privacy-preserving methods should be developed for the comfort of the people monitored [3] and ensure reliable, secure communication and continued functionality while preserving patients' safety, confidentiality, and data integrity. There is nearly universal agreement on the importance of security for personal health information and electronic health records, but there is still a disagreement over the security requirements for medical devices [2].

Security must be considered in early design phases [1]. Some approaches have explored the feasibility of protecting an implantable device from privacy attacks by implementing security mechanisms entirely on an external device [129] or by encrypting data [1‘3]. Moreover, in an effort to ensure security, personal authorizations and authentication have been proposed [20]. Therefore, Information and Communication Technologies (ICT) systems must facilitate the re-design of the current processes of care and follow up through the provision of services that enable the correct management of the patients within the healthcare organizations [130].

New and emerging technologies upset established moral norms by bringing to surface issues which were not previously open for discussion [131]. Argumentative patterns in this field are now known as NEST-Ethics (New & Emerging Science and Technology Ethics) [132].

5. Conclusions.

After the revision of the current state-of-the-art of the implantable multi-sensor devices, the authors propose a generic multipurpose in vivo implantable biomedical device capable of detecting several threshold values for targeted concentrations. As a result, an integrated front-end architecture for in vivo customized detection is embedded within an implantable device with a generic array of nanosensors combining cell clinic solutions as a lab-on-a-chip and electrical sensing. The key point in this new conception is that, instead of defining a particular architecture of the implantable device for each sensor, the new approach introduces the design and use of a general architecture that will require minor modifications for the final customized implantable device that could be suitable for a set of specific applications.

Given the speed with which chronic diseases are increasing and the aging of the world population, the improvements that are possible with new theranostics techniques could have a great impact on the wellbeing and quality of life of the whole society while suitable biomedical devices are designed to reach a huge market over the next few years. Thus, a successful research, development, innovation and technology transfer may be fostered in a particular scenario typified by the convergence of technologies and disciplines, as well as by the combination of several KETs allowing the pilot lines and commercialization of cutting-edge devices embedding implantable sensors. Amongst all KETs, in this blending of technologies, nanotechnology seems to have a great impact, enabling new advantages in medical diagnostic or therapeutic devices, from the use of nanomaterials, in the development of nano-biosensors, by using the engineering of surfaces in order to improve the sensitivity of an electrode or its biocompatibility, and using nanoparticles from a therapeutic or diagnostic point of view, allowing modulation of treatment to particular targets within the human body, and ensuring delivery in an optimal way for a specific patient.

Although the case study reported in this paper is complex because it involves multiple organizations and sources of data, it contributes to extending experience to the most recent developments and practices on implantable sensors. The next step involves the development of a configurable application-specific integrated circuit (ASIC) working with a multiplexed array of nanobiosensors designed to be reactive for a set of target agents (enzymes, viruses, molecules, chemical elements, molecules, etc. ). In this way, multiple sensors of the array can be used for one specific target, while other arrays can be prepared for the other targets, while also seeking a redundant response. As a result, a customized panel of biomarkers will be ready to be embedded into the bioimplantable medical device: each array will be used to detect a specific type of target, and the multiplexed system will be used to analyze each array focusing on a particular target. Then, top down approaches using nanoengineering and nanofabrication and bottom up approaches using supramolecular chemistry can produce novel diagnostics which will increasingly focus on delivering a personalized solution based on a real time analysis of array data, and where appropriate, applying this decision to deliver an automated therapy (theranostics).

The modular standard lab-on-a-chip approach introduced in this work may adapt the sensors in a quick, efficient and reliable way. Moreover, the system described in this paper must be tested before its implementation in a human being, and a POC platform would be designed for this purpose. The multi-parametric configurable implantable biochip system would be placed as a plug-and-play device. Moreover, it is needed to place a chip in the electronics module for the generation of the CV signals to check the sensors after their functionalization. Communications and powering will follow the same wireless approach as the implantable device. Once the performance of the sensors has been certified and cleaned-up, the implantable system will be suitable for being placed in the patient. This concept of programmable platform could be adopted in the design of a POC external device.

On the other hand, despite the somewhat limited availability of information discussing the safety of implantable sensors, the case study presented in this paper is a clear demonstration of how to take into account biocompatibility challenges and ethical concerns to foster the development of new bioimplantable medical devices. At this point, the bonds between the science community, hospitals, industry and citizens need to be strengthened with the aim of enhancing biomedical research on implantable sensors and its commercialization. Doubtless, biomedical devices represent a strategic gamble for the future of scientific and technological policy areas as they seek accelerated economic growth within the knowledge-based society and confront the new scientific and market challenges presented by the nano-enabled implantable biomedical devices.

Finally, the present and future of the implantable devices goes beyond these objectives and research challenges. There is a great transformation in medical diagnostics and the blend of the different KETs for the integration and commercialization of these devices should follow a standardization process to propel them in a Moore's Law trajectory as happened with the microelectronics revolution.

Author Contributions.

Esteve Juanola-Feliu is the corresponding author of this paper and its main contribution is to introduce a new conception of personalized implantable sensors that will require minor modifications for the final customized implant. Moreover, he promotes to bridge the gap from the lab to the market fostering innovation and technology transfer in the fields of nanobiotechnologies and biomedical devices.

Pere Miribel-Català and Jordi Colomer-Farrarons contribute with the state-of-the-art of the implantable devices as well as designing an integrated front-end architecture for in vivo customized detection and the corresponding electronic modules and measurements.

Cristina Páez-Avilés contributes with the market research, the multi-KET approach and ethics concerns for implantable sensors.

Manel González-Piñero contributes with the innovation process and value enhancement of the biomedical research focused on implantable sensors.

Josep Samitier is the director of the research group and supports applied research and innovation on biomedical engineering.

Genomic diversity and adaptation of Salmonella enterica serovar Typhimurium from analysis of six genomes of different phage types.

خلفية.

Salmonella enterica serovar Typhimurium (or simply Typhimurium) is the most common serovar in both human infections and farm animals in Australia and many other countries. Typhimurium is a broad host range serovar but has also evolved into host-adapted variants (i. e. isolated from a particular host such as pigeons). Six Typhimurium strains of different phage types (defined by patterns of susceptibility to lysis by a set of bacteriophages) were analysed using Illumina high-throughput genome sequencing.

Variations between strains were mainly due to single nucleotide polymorphisms (SNPs) with an average of 611 SNPs per strain, ranging from 391 SNPs to 922 SNPs. There were seven insertions/deletions (indels) involving whole or partial gene deletions, four inactivation events due to IS 200 insertion and 15 pseudogenes due to early termination. Four of these inactivated or deleted genes may be virulence related. Nine prophage or prophage remnants were identified in the six strains. Gifsy-1, Gifsy-2 and the sopE2 and sspH2 phage remnants were present in all six genomes while Fels-1, Fels-2, ST64B, ST104 and CP4-57 were variably present. Four strains carried the 90-kb plasmid pSLT which contains several known virulence genes. However, two strains were found to lack the plasmid. In addition, one strain had a novel plasmid similar to Typhi strain CT18 plasmid pHCM2.

استنتاج.

The genome data suggest that variations between strains were mainly due to accumulation of SNPs, some of which resulted in gene inactivation. Unique genetic elements that were common between host-adapted phage types were not found. This study advanced our understanding on the evolution and adaptation of Typhimurium at genomic level.

خلفية.

Salmonella enterica serovar Typhimurium is one of the leading causes of Salmonella - related gastroenteritis in humans. The Anderson phage typing scheme [1], in which Typhimurium is divided into subtypes based on phenotypic variation, resulted from the susceptibility or resistance to a set of bacteriophages, has been used for epidemiological typing for the past 40 years. The success of phage typing has been well documented in the tracking of epidemiological phage types such as DT204 in the early 1970s, and recently, the epidemic strain DT104 causing outbreaks worldwide [2]. Typhimurium is a broad host range serovar but has also evolved into host-adapted variants with some phage types being more commonly isolated from particular hosts. For example, DT2 is commonly isolated from pigeons and is associated with systemic disease in pigeons [3]. Host adaptation ensures the circulation within an animal population, and this process may have evolved through the acquisition of virulence determinants and/or loss of gene functions [4].

In Australia, three phage types were found to be predominantly isolated from human infections as well as in animals based on surveillance data from 1996 to 2018 [5]. DT135 has been most prevalent, causing 20-27% of Typhimurium infections in the past 10 years, and is clearly established in Australia as an endemic phage type infecting humans. DT9 is less frequent than DT135 in human infections but is the most frequent phage type in farm animals, with almost twice the frequency. DT170/108 has been increasing steadily over recent years and became the most frequent phage type in 2004, surpassing DT135 [5].

We previously addressed the origins and relationships of the common phage types in Australia using single nucleotide polymorphisms (SNPs) as molecular markers [6]. SNPs discovered by amplified fragment length polymorphism analysis were used to determine the genetic relationship of 46 Typhimurium isolates from nine phage types [7]. SNP typing was later extended in our recent study to incorporate additional SNPs obtained from comparison of five Typhimurium genomes (DT2, LT2, SL1344, > D23580 and NCTC13348) [6]. A total of 44 SNPs were able to resolve 221 Typhimurium isolates with 45 phage types into four major clusters (SNP clusters I to IV). However, the SNPs used clearly still have limited discriminatory power. There were SNP profiles which contained many different phage types. Phage types that are prevalent in Australia, including DT9, DT135 and DT197, were clustered with other phage types. Due to SNP discovery bias [8], more SNPs from strains representing the diversity within the SNP profiles or phage types will be required to increase the resolution of SNP based typing.

Genome variations have been observed in studies comparing whole genome sequences of Typhimurium strains LT2 (DT4), SL1344 (DT44), NCTC13348 (DT104), and > D23480 (unknown phage type), with strain specific pseudogenes and SNPs found within each genome [9]. Mobile elements such as prophages, transposons, plasmids and insertions sequences may also vary among Typhimurium genomes. The aims of this study were to use comparative genomics to identify the genetic diversity between multiple prevalent phage types and try to begin to elucidate the genetic basis for the predominance of certain phage types and host adaptation.

Bacterial strains and genomic DNA isolation.

Six strains were selected to represent the spectrum of Typhimurium diversity (Table  1 ) based on a previous SNP typing study using 44 SNP markers [7]. The selection criteria were based on SNP profiles as well as the prevalence of certain phage types from epidemiology data collected from the National Enteric Pathogen Surveillance Service (NEPSS) [5,10]. A DT99 isolate (host-adapted to pigeons) was selected for comparison against other Typhimurium isolates to identify potential genetic factors involved in host adaptation. Genomic DNA from each strain was extracted using the phenol/chloroform method as described previously [11].

DNA sequencing and assembly.

DNA libraries were prepared with insert size of 500 bp and were sequenced using the Illumina Genome Analyzer (Illumina). We used 2 × 50 bp paired end sequencing. Contigs were assembled using the Short Oligonucleotide Analysis package (SOAP) (version 1.04) [12]. SOAPdenovo settings were set with the following parameters: K value = 31, ‘d = 1 and D = 1 to generate scaffolds. A K value of 31 gave the best N50 contig size. Large scaffolds and short contigs generated by SOAPdenovo were aligned to the Typhimurium LT2 genome ( > NC_003197) using progressiveMauve [13].

Identification of SNPs.

Mapping of reads against the Typhimurium LT2 genome was performed using the Burrows-Wheeler alignment (BWA) tool (version 0.5.8) [14]. The output generated lists including the number of Illumina reads covering each nucleotide position, which corresponds to the reference genome. A custom script was used to extract SNPs according to the position on the reference genome and the number of reads covering the region containing the SNP. Some SNPs could result from errors in mapping or sequencing. Therefore, further filtering was performed.

A cutoff of 20 reads covering the SNP site was used initially to remove SNP sites with low coverage. We also used SOAPdenovo to assemble reads into contigs and then compared with the LT2 genome to identify SNPs. de novo assembly was done using quality trimmed reads. This may have reduced the number of SNPs. de novo assembly eliminated the problem with reads that may be mapped to spurious positions (mostly repeats or homologous regions) with mismatches being called SNPs. For SOAPdenovo assembly, reads were trimmed after the first base falling below Q7. The read was only excluded if the length of reads was 17 bases after the trimming. For BWA mapping, no filtering of reads was performed.

SNPs identified by both methods were compared. These common SNPs were manually inspected using SAMtools (version 0.1.7) [15] and its in-built function, Tview, for visualising the mapping of reads at each SNP position. SNPs identified from BWA mapping were further filtered using SAMtools by SNP quality. Any SNPs with quality score of less than 20 were removed.

SAMtools were used to manually confirm all SNPs for our initial analysis of one genome. We found a consistent pattern where SNPs were in fact sequencing errors when the region was covered only by ends of reads which is known to have poorer quality. For SNP sites with heterogeneous reads (i. e. at least two bases were called at the same site from different reads), the majority of the SNPs were genuine if the SNP was supported by ≥70% of the reads. A small proportion of SNP calls were genuine for those falling between 30% and 70%. None of the SNPs was genuine if less than 30% of reads supported the SNP. In case we removed genuine SNPs of lower than 20X coverage, we inspected SNP sites between >10 and <20 reads coverage and rescued genuine SNPs and added to the final set of SNPs. These genuine SNPs with lower than 20X coverage generally had 100% support for a SNP. Non-genuine SNPs were typically located at the ends of the reads and visual inspection identified them with relatively low subjectivity.

Distribution of insertions and deletions.

Insertions and deletions (indels) were identified using the mapping data from BWA [14]. The distance between the paired ends of a read were first calculated by mapping them to the reference genome. Any pairs with distance larger than the average insert size of 500 bp potentially contain a deletion in the newly sequenced genome. There were at least 10 fragments (paired end reads) to identify a deletion. The regions containing the potential deletion were examined using the Tview function in SAMtools [15] to locate the approximate breakpoint of the deletion and determine the number of reads covering the region up to the breakpoint with at least 20X coverage for the confirmed deletions. It should be noted this approach cannot identify small indels. We only looked for deletions of at least 500 bp in the new genome. This approach cannot identify large insertions in the new genome either. Potential indel events were further compared to other publicly available genomes that were found to be closely related from the SNP-based phylogeny to determine whether they were present in the other genomes [6].

Identification of new IS insertions were done using a similar method. Paired end reads with one end mapping to an existing IS location in the reference genome while the other end mapping to a distance location. The insertion point was determined visually based on the typical pattern of abrupt end of reads mapping as no overlapping reads can be found at the insertion point. We did not determine the precise location of the insertion using reads that contain part of unique sequence and part of IS sequence.

Identification and annotation of unique sequences.

Using progressiveMauve [13], some contigs were not able to be aligned. These unaligned sequences were re-aligned using BLASTn against reference Typhimurium genome strain LT2 to confirm whether they were duplicated sequences or unique regions. After duplicated sequences were identified and removed, contigs that did not belong to LT2 were compared again using BLASTn against the GenBank non-redundant nucleotide collection database to determine their homologues.

Phylogenetic analysis.

Maximum parsimony was done using PAUP [21] with heuristic search based on tree bisection and reconnection (TBR) swap method. S. enterica serovars Enteritidis and Choleraesuis strains were used as an outgroup. Outgroup genomes were aligned using progressiveMauve to the LT2 reference genome. A custom script was used to extract the nucleotide for each outgroup genome at the corresponding SNP containing locations.

Results and discussion.

Selection of strains and genome sequencing.

Two isolates were selected from SNP cluster I from our previous study [6], including L945 (DT108) and L927 (DT12a). L945 is a DT108 isolate but in Australia, DT108 is also known as DT170. This phage type contributes to approximately 15% of Typhimurium infections in Australia [5]. The phage type DT12a was a prevalent phage type in Australia but has decreased in recent years [5]. However, DT12a was reported with increasing infection during poultry surveillance in the neighbouring New Zealand [22] and multidrug resistance as reported by Lawson et al. [23]. L847 (DT197) was selected to represent SNP cluster II and is one of the most prevalent phage types [24]. There were three strains selected for SNP cluster III, L852 (DT135a), L904 (DT9) and L796 (DT99). DT135 and DT9 are the two most prevalent phage types in Australia. L852 is a DT135a strain, a variant of DT135 which belongs to the same SNP profile as other DT135 strains. A DT135a strain was selected over DT135 since it has been increasing in frequency in recent years in Australia. L796, a DT99 isolate, was selected as a representative of host adapted phage type. DT99 has been commonly isolated from pigeons [25]. The genome data of this strain provides a comparison with DT2, a phage type adapted to pigeons which is currently being sequenced by the Wellcome Trust Sanger Institute (sanger. ac. uk/resources/downloads/bacteria/salmonella. html). All except DT99 selected in this study are broad host range phage types.

The average number of reads (50 bp) generated per genome was.

9,200,000 and the coverage depth on average for all genomes was.

75X, with the lowest coverage of 63X (Table  1 ). Coverage of the reads against the LT2 reference genome ranged from 1 to 662 reads per site. Those with low coverage are most likely reads with multiple sequencing errors aligned in the wrong position. Genome coverage based on the LT2 reference was approximately 95% on average for the six genomes, with 96% coverage for strains L796 (DT99), L847 (DT197), L852 (DT135a) and L904 (DT9), 91% for L945 (DT108) and 98% for L927 (DT12a). The difference in coverage was likely due to genome sequences present in the LT2 reference but absent from the strains sequenced. Mapping was also performed on the plasmid pSLT associated with the LT2 reference, which is described in detail below.

Single nucleotide polymorphisms in Typhimurium strains.

We used two approaches to identify SNPs: mapping to the reference genome and de novo assembly. We first used BWA [14] to map reads against the LT2 genome which generated a large set of possible SNP sites (Table  2 ). For each potential SNP, a cutoff of 20 reads covering the SNP was required in order to exclude SNPs generated from incorrect mapping or sequencing errors. After filtering the SNP sites with low coverage, the average number of SNPs for the six strains was 2,631 SNPs. Strain L796 (DT99) contained the most number of SNPs with 3,323 SNPs identified. We then used SOAPdenovo to assemble reads into contigs and then compared with the LT2 genome to identify SNPs. Four genomes, L796 (DT99), L847 (DT197), L945 (DT108) and L927 (DT12a) had lower numbers of SNPs than direct mapping of reads using BWA. SNPs identified by both methods were further filtered by SNP quality and SNPs with a quality score of less than 20 were also removed to derive a final set of SNPs. There were on average 611 SNPs, ranging from 391 SNPs in L945 (DT108) to 922 SNPs in L796 (DT99). The confirmed number of SNPs was approximately a quarter of the original number of SNPs identified and this was due to mismatched reads alignments and sequencing errors.

The SNPs were classified into four categories: non-synonymous (nsSNP), synonymous (sSNP), intergenic (IG) and single base indels. IG SNPs on average accounted for approximately 17.5% of the total number of SNPs in each strain. The average percentage of sSNP for each strain was 36.2% with L904 (DT9) having the lowest ratio of sSNPs (32.5%) while the nsSNPs ranged from 42.1% to 46.8% with an average of 43.9%. Single base indels accounted for only 2% of the SNPs. L796 (DT99) had the highest number of indels (32 indels) representing 3.5% of SNPs, while L927 (DT12a) had the lowest with eight indels which accounted for 1.5% of all SNPs. The ratio between sSNPs and nsSNPs ranged from 0.70 to 0.89.

Genome tree based on maximum parsimony analysis.

The SNP data from the six isolates were compared with those from seven publicly available Typhimurium genomes: LT2, SL1344, > D23580, 14028 S, T000240 and DT104 and an incomplete DT2 genome. A maximum parsimony (MP) tree (Figure  1 ) was constructed using 3,368 SNPs with two other serovars, Choleraesuis strain SC-B67 and Enteritidis PT4 strain NCTC 13349, as the outgroup. Only one MP tree was generated. The homoplasy index (HI) was 0.112, indicating the presence of parallel or reverse changes as discussed below. The HI was comparable to what was found previously using SNP typing (HI = 0.17). SNPs identified from prophages were not used to generate the MP tree as the addition of prophage SNPs resulted in a HI of 0.212. External branches contained more SNPs than internal branches.

In the MP tree, the Typhimurium strains were distributed into three clusters designated as genome cluster (GC) I, GC II and GC III. Strain L927 (DT12a) was found to have diverged the earliest since it was closest to the outgroup and shared most recent common ancestor with the three clusters which were grouped together and supported by 97 SNPs. GC II and GC III were grouped together and supported by 64 SNPs. GC I contained three isolates L945 (DT108), T000240 and LT2 and was supported by 56 SNPs. GC II contained four publicly available genomes DT2, 14028 S, SL1344, > D23580, and three NGS strains; L852 (DT135a), L904 (DT9) and L796 (DT99). GC II was well separated from GC III and was supported by 203 SNPs. GC III contained two isolates, DT104 and L847 (DT197) and was supported by 32 SNPs. The strain specific SNPs for L796 (DT99), L847 (DT197), L852 (DT135a), L904 (DT9), L927 (DT12a) and L945 (DT108) were 378, 274, 96, 202, 197 and 194, respectively. Amongst the publicly available genomes, DT2 had a large number of strain specific SNPs (683 SNPs) while the strain specific SNPs for the remaining six genomes (LT2, T000240, SL1344, 14028S, DT104 and > D23580) ranged from 71 to 151.

In this study, the HI was greater than 0 which suggests that some SNPs had conflicting phylogenetic signals. The SNPs were mapped onto the internal and external branches in the MP tree. There were 33 SNPs present in multiple internal branches, indicating reverse/parallel changes which were likely to be resulted from recombination between lineages. Of these, 28 were IG SNPs. For the external branches, the distribution of SNPs was used as an indicator of recombination within a lineage. It has been previously suggested that recombination event constitutes the presence of three or more substitution events within a 1 kb region [26]. There were 47, 31, 30, 10, 14 and 18 SNPs, which resulted in 34, 17, 24, 6, 10 and 13 potential recombinational segments in strains L796, L904, L847, L852, L927 and L945, respectively. The recombination to mutation rate was similar in L796, L847, L904 and L945 with approximately 5% of their SNPs resulting from recombinational events while L852 had the lowest with only 1.9%. Despite the presence of parallel/reverse mutations, the resulting phylogeny was generally consistent with our previous SNP typing study. This suggests that the extent of recombination has not yet distorted the evolutionary relationships among the strains and may not play a major role in driving the genetic diversity within this serovar.

Insertion sequences.

Insertion sequences (IS) play an important role in bacterial evolution as transposition can potentially inactivate a gene [27]. S. enterica is known to carry three insertion sequences, IS 1, IS 3 and IS 200 [28]. IS 1 is rarely detected in Typhimurium [28] and was not found among the six NGS strains. IS 3 was previously found in a high proportion of Typhimurium isolates included in Salmonella reference collection A (SARA) [29]. However, only a single copy of IS 3 sequence was found at the same location in the six strains sequenced. IS 200 was the only IS commonly found in the genome strains.

There were 32 different IS 200 locations found among the 13 Typhimurium genomes, with 27 located in IG regions (Figure  2 ) and five in genic regions. The genic insertions are described in the gene disrupting events section below. Several of the IG IS 200 locations were found in the same location in multiple genomes. The reference genome LT2 contained six copies of IS 200 in IG regions, which were designated as IS 200 _1 ‘ IS 200 _6. The other IS 200 copies are assigned as IS 200 _7 to IS 200 _27. Fifteen of these IS 200 locations (IS 200 _9 ‘ IS 200 _20, IS 200 _23 and IS 200 _26) occurred only in a single strain as shown in Figure  2 - The average number of copies of IG IS200 was seven ranging from four to 10 copies. IS 200 _1 and IS 200 _2 were likely to have been gained by the most recent common ancestor of GC I, GC II and GC III since L927 (DT12a) does not contain either IS. Additionally, IS 200 _2 was likely to have been lost by L904 (DT9) and L852 (DT135a). Similarly, IS 200 _24 and IS 200 _25 were likely to have been gained by the most recent common ancestor of GC II as all except one strain contained IS 200 _24 and all contained IS 200 _25 (Figure  2 ).

Gene disruption events.

Five IS insertions were found within a gene. The IS 200 insertion at the accA gene (L945 (DT108)) and brnQ (DT104) occurred in a single strain. The insertions at dacB , icdA and rnd were shared amongst two or more strains. The disruption of rnd appeared to have occurred early during the divergence of GC II as all strains belonging to this cluster had the IS 200 insertions. The insertions of IS 200 into dacB and icdA occurred three and two times respectively. It appeared to be independent for both cases as the gene disruptions were present in strains from different clusters (Figure  2 ). We did not determine whether the independent insertions occurred at the exact same site for each gene as potential hotspots for IS insertion.

The insertion into icdA , found in L945 and 14028 S, may have an advantageous effect, with icd mutants known to resist low levels of nalidixic acid [30]. The non-functioning icd gene results in the lack of citrate synthase activity, allowing the accumulation of citrate which has an unexplained effect on nalidixic acid resistance.

AccA codes for carboxyltransferase subunit and can be inhibited by pseudopeptide pyrrolidinedione antibiotics such as moiramide B. Pyrrolidinedione resistant strains of E. coli and Staphylococcus aureus have been found to contain mutations in the subunits of AccA [31]. The IS 200 disruption of accA in L945 (DT108) may be related to antimicrobial resistance.

Another inactivated gene dacB was found in strains L945 (DT108) and L847 (DT197). This gene encodes a penicillin-binding protein 4 (PBP4) which functions as a trap target for β-lactams [32]. Interestingly, the inactivation of nonessential dacB - coded PBP4 triggers overproduction of β-lactamase AmpC and the specific activation of the CreBC (BlrAB) two-component regulator leading to a high level of β-lactam resistance [33].

Prophage insertions and deletions.

Typhimurium strains harbor many prophages [34-40]. The pairwise alignments of the NGS strains against the reference LT2 genome were used to determine the presence and absence of prophages. In total, nine prophages or prophage remnants including Fels-1, Fels-2, Gifsy-1, Gifsy-2, Gifsy-3, ST104, CP4-57, ST64B and SopEΦ were found in the six NGS genomes (Table  3 ).

Prophages Gifsy-1, Gifsy-2 and the sopE2 and sspH2 phage remnants were present in all six genomes. They contain virulence genes commonly associated with type III effector proteins that are injected by the bacterium through type III secretion [41], and thus are important for virulence.

Fels-1 and Fels-2 were only found in strains, L847 (DT197) and L904 (DT9), respectively. The sparse presence of these prophages was not surprising as Fels-1 and Fels-2 are commonly absent in Typhimurium strains [42]. Fels-1 codes for virulence factors nanH and sodCIII [41] while Fels-2 harbours the gene abiU of unknown function [41]. ST104 prophage harbours artAB which codes for a putative toxin. It is often present in epidemic multiple drug resistant DT104 strains [43]. Interestingly, this prophage was also found in L927 (DT12a).

ST64B codes a virulence factor similar to Ssek NleB type III secreted effector proteins [44]. This prophage was present in four NGS strains, L847 (DT197), L852 (DT135a), L904 (DT9), and L945 (DT108), which were from both GC I and GC II. ST64B was also found in genome strains, SL1344 and 14028 S, but the prophage is defective in these strains due to a frameshift mutation in the open reading frame (ORF) SB21, which leads to the inability to produce infectious virions [45]. The same frameshift mutation in SB21 was found in L852 (DT135a) and L904 (DT9). Since SL1344, 14028 S, L852 (DT135a) and L904 (DT9) all belonged to GC II, the frameshift mutation may be an ancestral event shared by these GC II strains.

Prophage CP4-57 controls phage excision during the biofilm development stage which in turn enhances the motility in the host and increases biofilm dispersal while reducing growth [46]. This prophage has been found in E. coli strains suggesting a co-evolution between the two species [46]. This prophage was only found in strain L852 (DT135a). Since DT135a is a prevalent DT, further studies are warranted to determine the role of CP4-57 in adaptation and DT135a prevalence. Biofilm formation and dispersal are likely to be important for environmental survival leading to prevalence [47,48].

The P2-like phage SopEΦ was notably absent in all six strains sequenced in this study. SopEΦ contains the virulence gene sopE , and when disrupted reduces invasiveness [49]. This prophage was found in epidemic Typhimurium strains of DT204 and DT49 [39] and in two published genome strains SL1344 and ST4/74.

Altogether, the results suggest that prophages may not be maintained in all Typhimurium genomes. On the other hand, other studies have shown that prophages can be easily transferred between strains, particularly if the prophages are integrated at a tRNA site, like the case of ST104 and ST64B [50]. Thus, prophages make an important contribution to the diversification of Typhimurium genomes. The analysis also highlights the important roles prophages may play in virulence and potential adaptation of Typhimurium.

Large indels.

Regions of > 500 bp that are present in the reference genome LT2 but absent in the NGS strains are identified as large indels. Seven large indels were found in the six NGS strains with sizes ranging from 558 bp to 1,992 bp (Table  4 ). Indel events were cross-referenced with genomes SL1344, > D23580, DT2 and 14028 S to see if they were carried on closely related isolates. Five indels were present in only one strain, with two indels shared in more than one strain. Deletion of STM2599 ( gipA ) was found in L927 (DT12a) and L847 (DT197) and the indel in STM0291 was found in L852 (DT135a) and SL1344. Three of the seven deletions encompassed a whole gene while the remaining four deletions were partial deletion of a gene leading to truncations of 15% to 37% of the gene. Truncations of more than 20% were treated as pseudogenes [51]. Thus, all, except STM0291 which only had a truncation of 15% of the gene, were likely to be nonfunctional. STM0291 is located in SPI-6 and codes for a RHS like protein. RHS family proteins have been known to be diverse [52]. Therefore, the deletion detected in L852 and SL1344 may be a variant of RHS.

Interestingly, the genes deleted or truncated appeared to have a role in virulence. gipA is required for survival in Peyer’s patches [40]. gipA mutants have been shown to be attenuated to some extent after oral infection in mice, but displayed the same level of virulence as the wild type if inoculated intraperitoneally. safA ( STM0299) is part of the saf fimbrial operon. safA mutants are attenuated in a pig model, but not in calves or chickens [53], and the same saf operon is not needed for virulence in mice [54,55]. sopA is used to alter host cell physiology and promote bacterial survival in host tissues [56]. sopA mutant has reduced Salmonella - induced polymorphonuclear leukocytes transepithelial migration [56].

Loss of these genes is expected to be disadvantageous to each corresponding strain, and may explain the differences in the ecology of several phage types. L847 (DT197) carries gipA deletion, sapA and sopA truncation. However, this phage type has increased in frequency in Australia in recent years, which argues against the importance of these genes for virulence in humans, although the increased detection of this phage type may be due to increased ability to colonise food animals, leading to increased exposure in humans. L927 (DT12a), also one of the most frequent phage types, carries gipA deletion and contains a truncated STM4534, a transcriptional regulator which regulates the phosphotransferase system and possibly other systems [57].

Gene disrupting mutations.

Apart from gene truncations due to partial deletion of a gene as described above, we identified 15 genes terminated earlier when compared to strain LT2, due to a stop codon (Table  5 ), leading to proteins that are >20% shorter and thus these genes were considered as pseudogenes. The distribution of these pseudogenes was mapped onto the SNP-based MP tree (data not shown). All, except one pseudogene (STM3745), was found only in a single strain as a single event. Earlier termination in STM3745 of unknown function resulted in 28% shorter protein and was observed in three strains of the same lineage, L796 (DT99), L904 (DT9) and > D23580, suggesting a common ancestral event.

L796 (DT99) had the most number of strain specific pseudogenes followed by strains L927 (DT12a), L847 (DT197), DT2 and L904 (DT9) with seven, three, three, two and one pseudogenes, respectively. It is interesting to note that L796 (DT99) had a higher number of disrupted genes as well as the highest number of SNPs. This strain also had a 27% shorter DNA polymerase II encoded by polB, which may have resulted in a mutator phenotype.

Several pseudogenes for example, napF and blc , if active, are involved in energy conversion and metabolic pathways. napF encodes a predicted 3Fe-4S iron sulfur protein [58]. NapF mutant causes a growth defect under anaerobic conditions on glycerol/nitrate medium but is not essential for the activity of periplasmic nitrate reductase [59]. Therefore, NapF does not have a direct role in nitrate reduction but contributes to energy conservation. In E. coli , blc promoter is expressed during stationary growth phase which is controlled by rpoS sigma factor, directing the expression of genes necessary for adaptation to low nutrients condition. Therefore, both napF and blc are important for conserving energy.

Only one pseudogene, sthB , may be involved in host-restriction. SthB , if active, codes for a fimbrial usher protein. Deletion of sthABCDE operon results in reduced caecal colonisation in mice [60]. Furthermore, sthB mutants in chicken hosts have reduced colonisation [55]. This pseudogene was only found in L796 (DT99). Since this strain is only commonly associated with pigeons, this gene may have an effect on host restriction.

Plasmid sequences.

Most Typhimurium strains including LT2 carry a 90-kb virulence plasmid, pSLT [61]. It contains many known virulence genes including spv ( Salmonella plasmid virulence), the pef (plasmid-encoded fimbriae) region, rck (resistance to complement killing), a homolog of dsbA (disulfide bond isomerase) and a homolog of the AraC family of transcriptional regulators [62-65]. The published genomes SL1344, DT2, > D23580 and 14028 S were all found to contain pSLT. In order to determine the presence of pSLT from the 6 strains sequenced, reads and contigs were mapped onto the LT2 pSLT sequence ( > NC_003277). Reads homologous to pSLT were found in strains L796 (DT99), L847 (DT197), L852 (DT135a) and L904 (DT9) with 96%, 96%, 86% and 90% coverage to pSLT, respectively.

Strains L945 (DT108) and L927 (DT12a) contained reads covering only 2.6% and 0.8% of the pSLT plasmid suggesting that both of these strains did not have pSLT. It is likely that L945 (DT108) has lost the plasmid as all other strains from GC I contained the plasmid. In contrast, it is not known whether L927 (DT12a) has lost the plasmid or the plasmid was only gained after the divergence from the L927 (DT12a) lineage.

L945 (DT108) contained additional contigs that were not able to be aligned with LT2 chromosomal sequence or pSLT plasmid sequence. These contigs were then searched against GenBank using BLASTn. Eighty eight contigs, ranging from 104 bp to 5,980 bp, from L945 (DT108) had the closest match, with 65% DNA sequence identity, to the cryptic plasmid pHCM2 of 106 kb belonging to Typhi strain CT18. Sequence homologous to repA was identified in one of the contigs suggesting that a novel plasmid was present in L945 (DT108).

Comparison of host adapted phage types: DT99 and DT2.

Phage types of DT99 and DT2 are commonly associated with pigeons [3] and the mechanism for host-adaptation in these two phage types remains unknown. A previous microarray study on DT2 and DT99 phage types found that the loss of genetic regions does not correlate with host-adaptation [66]. The DT2 strain and L796 (DT99) were well separated within GC II and adaptation must have occurred independently. Comparison of their genomes did not identify any additional sequences that may contribute to host adaptation. There were few indels found common to both DT2 and L796 (DT99) that were not found in other genomes. A region between STM1555 to STM1559 was absent, which was previously reported [66]. This region encodes several proteins of putative functions including a transcriptional regulator; Na + /H + antiporter; an aminotransferase; glycogen synthesis protein glgX [67] and glycosyl hydrolase. It is not clear whether the absence of this region is important for host-adaptation in pigeons since it is also absent in NCTC 13348 (DT104), a broad host strain. Other genetic elements commonly absent in both DT2 and L796 (DT99) were the Fels-1 and Fels-2 prophages. Again, both of these prophages were also absent in many of the other Typhimurium isolates.

Studies have shown that adaptation could be resulted from changes as small as one SNP, which can result in either increased or reduced virulence in animal models [68,69]. For example, an rpoS mutant LT2 strain has reduced virulence in mouse models [69]. Similarly, a nsSNP on fimH has been shown to improve the bacterial adhesion of serovar Enteritidis to chicken leukocytes [70]. There were no SNPs in either of these genes in strains L796 (DT99) and DT2. nsSNPs in ycjF were found in both L796 (DT99) and DT2, although the SNP locations differed between the two, at codons 301 and 324 for L796 (DT99) and codon 181 for DT2. ycjF codes for a hypothetical protein and its homolog in E. coli is essential for virulence in vivo in a mouse septicaemia model [71]. Other than that, there were no SNPs that were only found in both L796 (DT99) and DT2.

Comparative genome analysis of DT99 and DT2 revealed few genetic features that are specific to these two host adapted phage types. Therefore, multiple factors are likely to have contributed to the adaptation to pigeons.

استنتاج.

Six diverse Typhimurium strains based on our previous SNP typing study were investigated at the genome level and compared to seven other publicly available genomes to determine genetic variations that may contribute to their prevalence and host-adaptations. Variations between these genomes largely resulted from accumulation of SNPs. These genome-wide SNPs were also used to establish the phylogenetic relationships of 13 genome strains. Despite the presence of parallel/reverse mutations, the resulting phylogeny was generally consistent with our previous SNP typing study [6]. Other variations included prophages, plasmids and IS elements. The pSLT virulence plasmid was detected in all except two strains, L927 (DT12a) and L945 (DT108). Interestingly, L927 (DT12a) contained a novel plasmid with some similarities to cryptic plasmid pHCM2 first reported in Typhi CT18.

There was evidence of genome degradation, including pseudogene formation and some large indels. The pseudogenes mainly resulted from earlier termination codons or IS 200 insertions which appeared mostly to be random. However, some IS 200 insertions may provide a selective advantage including insertions in icdA , accA and dacB , all of which are related to antibiotic resistance.

Comparison of two host-adapted Typhimurium phage types, L796 (DT99) and DT2, did not reveal any unique genetic elements between them. SNP-based phylogeny grouped these strains together in GC II but they were clearly of separate lineages. This suggests that host-adaptation is a result of convergent evolution. However, factors contributing to the prevalence and host-adaptation in Typhimurium remain to be uncovered. In conclusion, genetic diversity within Typhimurium is mainly due to accumulation of SNPs, some of which led to pseudogenes. Unique genetic elements that were common between host-adapted phage types were not found.

Competing interests.

The authors declare that they have no competing interests.

Authors’ contributions.

Experimental work and data collection were carried out by SP. SP, SO, RL, LF, BL and PRR contributed to data analysis and interpretation. The study was conceived and designed by RL and LW. The manuscript was drafted by SP, SO and RL. All authors have read and approved the final manuscript.

Acknowledgment.

This study was supported by a National Health and Medical Research Council project grant. Stanley Pang was supported by Australian-China research council fellowship for a six month visit to Nankai University.

Exon-primed intron-crossing (EPIC) PCR markers of Helicoverpa armigera (Lepidoptera: Noctuidae)

Applying microsatellite DNA markers in population genetic studies of the pest moth Helicoverpa armigera is subject to numerous technical problems, such as the high frequency of null alleles, occurrence of size homoplasy, presence of multiple copies of flanking sequence in the genome and the lack of PCR amplification robustness between populations. To overcome these difficulties, we developed exon-primed intron-crossing (EPIC) nuclear DNA markers for H. armigera based on ribosomal protein (Rp) and the Dopa Decarboxylase (DDC) genes and sequenced alleles showing length polymorphisms. Allele length polymorphisms were usually from random indels (insertions or deletions) within introns, although variation of short dinucleotide DNA repeat units was also detected. Mapping crosses demonstrated Mendelian inheritance patterns for these EPIC markers and the absence of both null alleles and allele ‘dropouts’. Three examples of allele size homoplasies due to indels were detected in EPIC markers RpL3, RpS6 and DDC, while sequencing of multiple individuals across 11 randomly selected alleles did not detect indel size homoplasies. The robustness of the EPIC-PCR markers was demonstrated by PCR amplification in the related species, H. zea , H. assulta and H. punctigera .

أرسل رسالة إلى أمين المكتبة أو المشرف لكي يوصي بإضافة هذه المجلة إلى مجموعة مؤسستك.

ISSN: 0007-4853 EISSN: 1475-2670 URL: /core/journals/bulletin-of-entomological-research.

المقاييس.

عرض النص الكامل.

تعكس مشاهدات النص الكامل عدد تنزيلات بدف وملفات بدف المرسلة إلى غوغل دريف و دروبوكس و كيندل و هتمل في النص الكامل.

آراء خلاصة.

تعكس المشاهدات الخلاصة عدد الزيارات إلى الصفحة المقصودة للصفحة.

* المشاهدات التي تم التقاطها على كامبريدج كور بين سبتمبر 2018 - 12 يناير 2018. سيتم تحديث هذه البيانات كل 24 ساعة.